> There's some speculation that all of these recent outages might be related to attempts at taking Genesis Mission online - the Trump / Oracle project to have AI tie into government and private services: https://en.wikipedia.org/wiki/Genesis_Mission
It’s an initiative to enable AI stuff across public and private domains, it’s not implausible that Trump DoE is e.g. strong-arming companies to rush some botched deployments.
Separately: People conspire. Take whatever level of manager/executive etc. meeting you’re privy to, the fudging around external transparency / comms that goes on there, and scale that out to a country. Now factor in current admin.
Despite the name, "conspiracy" as in "conspiracy theory" isn't about the conspiracy per se. It's about the sort of handwavey, vibes-based thinking used in certain common arguments that also involve conspiracies. The issue I have with this claim isn't the conspiracy part, it's the complete lack of any actual explanation or any reason to believe these things are connected beyond the timing and some vague feelings. Those just happen to be commonly found in conspiracy-based claims such as this one.
The boundary between public and private is not as discrete as you might suspect.
As for conspiratorial, many explanations fit the (highly limited in circulation) facts. There is no basis for demoting any explanation as ‘conspiratorial’.
What about the boundary between public scientific research and private IT infrastructure? I don't even understand what's being proposed here. 1. A scientific AI system comes online. 2. dot-dot-dot. 3. Outlook goes down. What's the dot-dot-dot? What even is the explanation, and in what way does it fit the facts?
I label it as conspiratorial because there's no explanation given for why there would be a connection, beyond the timing and some vague handwaving about a government project with an interesting name.
Also the linked Substack is very, very obviously a conspiracy theory nut. I instantly got that vibe clicking through the provided link, and poking around reinforces that strongly. For example, in https://substack.com/home/post/p-197154724 she draws a line between Genesis Mission and an electronic health records program called MHS Genesis on the basis of nothing but the name. And the supposed connection is that Genesis Mission will train AI on the data from MHS Genesis. Which makes very little sense. How useful would electronic health records be as training data? This question is not explored at all. It's just taken as a given that this would be a useful thing for the government. But that's just the beginning. She then pivots to saying that this massive supercomputer "where the body becomes the battlefield" is being built under the demolished East Wing of the White House. That's lip-strummingly crazy. That's "Tesla was killed because he invented free energy at Wardenclyffe" level nonsense. That's the sort of thing that acts as evidence against any claim it's near.
I had issues transferring Apple Cash from my wife's phone to mine. Tried a couple of times but it didn't work and I gave up. Apple's system status says that service is working fine. Weird!
Yes, Apple runs a multicloud environment. They've been moving more of their stuff to their own DCs but iCloud has always been one of the largest multi-cloud customers.
If you read the breakdown its actually the opposite. They value their data security more so they keep it on their own servers, where as services like Apple Music and Maps require huge varying amounts of compute which is much easier to scale in realtime on AWS infrastructure.
We are making updates to a “log in with apple” flow on our website.
We were testing this and it suddenly stopped working. (Both the new undeployed version but also the production version which hasn’t changed in weeks.) Seems connected to the outage.
Just got back from the Apple Store. They confirmed that everything was down, they could not process orders or returns, and new iPhones that people bought weren't able to be activated.
Honestly I was surprised Apple stayed up during the Microsoft outage as it affected so many interconnected services and infrastructure. I presume that the Cloud providers are shifting some of the workload around and caused other critical services of their customers to fail.
Interesting. This is the first I've heard of broad service interruption at Apple in a while, as I can't even remember the last time. Do you suspect something about puppet or their overall security posture to be implicated in this?
Hell they were more stable than all the other clouds so far. Maybe a giant cluster of redhat computers managed like 2007 is actually pretty robust compared to super kube control plane v42 with 100 add-ons and layers thrown in on top.
Production servers should be hardened, immutable, and remotely attestable appliances. Everyone should be able to prove every line of code in memory at all times, with access to secrets gated on signed quorum approval from multiple reviewers. They should not have dev tools, ssh, a shell, and certainly not capable of blindly modifying themselves at runtime if a single admin of a Puppet control server asks them to.
Such tactics mean a single compromised engineer laptop means a compromise of production, and thus a data breach is all but inevitable.
I would disqualify anyone seriously suggesting Salt, Puppet, Ansible or otherwise advocating for live-mutating production servers in a sysadmin interview today.
Talos Linux, Stagex, and my own operating systems I have custom made for various past employers.
Custom operating systems are not scary. Make a UKI so your entire OS is just a kernel, init, and a target daemon payload. Treat servers like firmware. The less code in memory the better.
That said, I am working on an general purpose hardened server OS to be released soon publicly covered in my last talk https://lance.dev/talks/bsm2026
I understand and subscribe to some of the things you say but:
> Such tactics mean a single compromised engineer laptop means a compromise of production, and thus a data breach is all but inevitable.
I hope you don't believe that an immutable image cannot be compromised. The commits done to the build system come from the same engineer laptops and, in a world where CVE are published ao frequently, said compromised image could be deployed largely in only a few minutes hours.
Also there is something to be said with lack of debugging tools. While I agree that reducing the surface of attack is the way to go, I am pretty sure the reason it takes so long for some of the big infra outages we have experienced from various cloud vendors comes from the inability of engineers to log on some systems, observe and potentially fix temporarily by hand to stop a race condition. There is often a race condition and serpent eating its own tail situation involved and logs might not even reach the monitoring infra anymore.
Going back to manage pets is not the way to go but I am not entirely certain removing ability for a subset of engineers to log on systems is exactly the way to go.
> I hope you don't believe that an immutable image cannot be compromised.
Obviously, but I will trust a unikernel with just kernel/init and an unprivileged sandboxed daemon to have dramatically less attack surface than a redhat system. Reducing attack surface is the job.
> The commits done to the build system come from the same engineer laptops and, in a world where CVE are published ao frequently, said compromised image could be deployed largely in only a few minutes hours.
That is what signed commits and signed reviews are for, and a CI that mandates both. I mentioned quorum control and this is the method I always use. I founded the only Linux distribution that does this.
> Also there is something to be said with lack of debugging tools.
In some cases one may wish to expose a custom, unprivileged, and sandboxed ssh daemon that provides basic logging and read only diagnostic tools to admins. This is usually fine. Hardening and observability can co exist. No need for a real ssh and huge traditional desktop linux shell interface attack surface.
But if I want to do major debugging with dangerous tools in prod, which should almost never be required, it should have to be with consent of another team member to add a -new- image to the fleet with those debug tools, temporarily.
This would only be required when the only way to debug is with production secrets, but the default should always be to reproduce on a duplicate system with dummy secrets.
That's not the entirely of their cloud exposure, however. A ton of their stuff runs in AWS, I wouldn't be surprised if they had things on other cloud providers as well.
The Microsoft outage wasn't an Azure outage anyway, it was specifically tied to Exchange Online. No one running their own stuff in Azure had any issues during that event
Mostly true, but not entirely. Azure Communication Services (ACS) was down alongside Exchange Online and the other Microsoft 365 applications. I verified this by testing upwards of a dozen times throughout the outage window. There was absolutely no indication made by Microsoft though that ACS was impacted in any way.
Amazon's internal direction on how they scale out AWS and build new features is heavily guided by "what will it do to one of our biggest customers that has a secret code name but everyone knows its Apple".
Apple is notoriously good at avoiding these widespread outages, by doing things independently regardless of industry trends. I am pretty sure they don't outsource their hosting to Microsoft.
Even if they had been heavy Azure users, the impact to Azure was basically 0. Only Exchange Online and connected services were down. Even OneDrive, SharePoint, and Teams worked in all the parts where they didn't interval with Exchange (e.g. Teams Calendar could show your pre-scheduled meetings and allow you to join them, just not create new meetings).
Isn’t Patch Tuesday a simpler explanation? Something deep needs to be patched and rebooted. Or a coordinated malware removal effort, where anything less than full reboots cause propagation.
Do you really think that when these events happen, every single computer is patched at the SAME time ?
Does that sound intelligent ?
Does this happen in other industries ? Does the electric company shutdown every single transformer/station/big machine i dont know the name and perform servicing at the SAME time ?
I was AT the Apple store buying a replacement airpod while the exchange outage was happening.
Apple was fine, but this poor old woman was at the genius bar trying to get some accounts connected. The genius at the Apple store was trying to explain the outage - and it just was NOT clicking for the woman. He must have said 4 variations of "Yes your accounts SHOULD connect, but Microsoft is having a major outage right now so they do not."
Yeah I was trying to log in and it couldn't send 2FA. I tried all the methods and considered doing a reset, but I realized that the site was loading weirdly and wondered if it was systemic. Glad I waited...hopefully this will be over soon.
Interesting that one doesn't say the app store but the dev one does.
Both the app store and dropbox (??) have been unable to connect for the last six hours for me while other apps on ios are fine. Dropbox's status website says everything is fine, and I can't understand how that could be affected by apple services being down.
Edit: Dropbox finally caught up and added "Sign in with Apple is not working as expected for some users" to their status page
That page, at the time of writing seems to be focused around iCloud and or "iCloud Account & Sign In." Many of the services, that page claims, are up if you're already signed in.
Anything that goes fast, accelerates, enhances, boosts, dazzles is, in some way, whether you realize it or not, whether you agree or not, is connected to AI.
Anything faulty, slow, outdated, error-prone, is clearly due to some human in some way shape or form. If there's an ostensible cause from AI, the actual problem is probably from a human not reviewing the code or following processes, or writing a bad prompt.
I complain about Apple often, but I very much appreciate that most of the features on their cloud-service apps (Reminders, Calendar) work natively with DAV services like Fastmail.
I started getting notifications that my Apple ID was logged out of iCloud, and I am not suffering for it, since I am not using iCloud for much.
If only they took the same approach to their in-store operations which are completely stopped even if you wanted to pay in cash. Makes them look incredibly unserious
Too many people take for granted that these open protocols exist for a reason.
On another note, using non-iCloud accounts with those apps often results in a “degraded" experience. Many of the features offered with iCloud accounts on those apps are missing when connected to a different service.
I was on the team that made a few of those apps when iCloud was being made. The apps already supported the open protocols from the 2000s, so iCloud was designed to also use them.
Them missing features you refer to are not some kind of anticompetetive move, it's that protocols don't advance to support these things.
The number of times I deleted my cookies, flushed the cache and removed any ad blockers, thinking I am not being allowed in because of one of those, before they reported the outage....oh well...another day
I've been getting the 'This Mac can’t connect to iCloud because of a problem with ...' error. I opened HN and immediately saw this. Better than a status page.
I think the grounding was a big deal, and the Microsoft thing lasted a day so it's a big deal. But would any of these other ones get so much attention otherwise? NSA? Apple Support (unless it's more than just support)? Even bank logins? I suspect some of the smaller ones are getting played up since we're looking for a pattern.
Apple Pay is usually an option on checkouts using Stripe's 'Payment Element', but atm it Apple Pay doesn't present itself as an option as is usually the case on Apple devices.
I just drove an hour for a pickup in-store and their whole system has been down for >1hr - no transactions possible at all
Edit: and apparently Apple takes appointments but has no methodology to cancel them due to catastrophic outages like this, nor do they have any ability to conduct non-digital transactions…typical tech
Some years ago I insisted on paying in cash at the Apple Store (something cheap like an accessory) and was met with visible disgust by the employees... as they struggled to figure out how to make change and under which table the cash drawer and receipt printer were hidden.
Working the front-end in retail and not being able to make change is wrong to me. Like a software developer who can't write code. Soon we will have cab drivers who can't drive.
In Switzerland recently I grabbed a coffee and the register spit out my change for me, while the young register person stood without having to do much of anything. I guess at some point they will flip the screen around and make it a self checkout pour your own coffee, then they don’t have to pay anybody!
Working the front-end in retail and not being able to make change is wrong to me. Like a software developer who can't write code. Soon we will have cab drivers who can't drive.
When I worked at an Apple Store in 2012-2014-ish we could take offline card payments and we had hidden cash drawers. I vaguely remember a new C-suite taking over the retail stores who got rid of both, though.
Brower or Ahrendts? The latter was a huge mistake by Tim Cook, as she was lured from Burberry, given $75million of RSUs that vested over 3 years, and she quit to the day when they vested. Thorough disgust among the inner sanctum engineers building things, myself included in that group.
agreed. from the cupertino internal labs point of view, we were like "huh?". brower proposed a bunch of draconian counterproductive very non apple helicoptering, and Ahrendts was just a ridiculous poser.
Typical over centralization, routing everything through data centers to mine our data to insert ads. Usual American "all eggs in one basket" turn left go fast like NASCAR.
The scale of the data center build outs are not technically necessary. It's all designed to make big tech too big to fail, scare aging pols deeply invested in big tech stocks; the real infinite money glitch...scare the bejeezus out of the aging political class and they'll bankrupt the world for you
The only SaaS app I really need is a Dropbox for offsite backup.
If shit goes to shit where we are people can ping remote contacts direct to fan out the news; proper point to point topology.
Instead we are forced into a hub and spoke topology and we are not the hubs
All other backend infrastructure could be whittled down to fiber and wireless network gear where upon inspection all ya can see is electrons or scrambled packets.
I don't know how often but if I saw it on the news absent anything else I'd file it under "things are crazy these days" rather than "things are crazy this week in particular".
I think such things are happening more and more year after year, which is newsworthy in and of itself. I'm not denying that. I'm not even denying that AI/cyber incidents on a big scale haven't happened and aren't coming. This is all newsworthy. I just have doubts about how people are interpreting the events of this particular week.
It's payment services with apple at the moment, and it looks like it's this 3rd party vendor they work with that might be the issue? https://status.stripe.com/
This is for Stripe merchants to accept payments with Apple pay, it's more like Stripe wrapping apple pay (and reporting the underlying Apple pay outage to their customers) than the reverse
I happened to be trying to update from Tahoe (macOS 26) to Golden Gate (macOS 27) and it kept saying "Your mac is up to date (macOS Tahoe 26.5.2)" instead of displaying the button to update. So I suspect macOS Software Update is also down/patchy.
While I agree and was thinking along the same lines, the converse can also be true: perhaps these other outages (NSA, Apple Support, United, etc.) _should_ have been noticed before?
the pattern is pretty easy: the proliferation of open-chinese models with cyber offensive + the already huge cyber crime sector that now only requires compute, and no more skills (which capped them before). Basically Prompt: "Make me as much money as possible" and offensive agents will (not being reliable from the start) start attacking any (critical) infrastructure for exploitation - that's why you see so much downtime and critical failures. I don't think it is state actors, since they will have to expect counter strikes.
Personally, I find this explanation plausible, but I would like to actually know the answer. A few of these certainly feel unusual, but between coincidence and AI fears, it's hard to know. I'd love if someone with credibility investigates.
For context, the Retail stores are also unable to do transactions, process repairs, or use any of their internal services for customers. They are essentially fully offline.
I slightly lean towards it not being an attack but, yes, we should all remain agnostic.
One reason is that if a company is under attack, 1) usually, you can tell you're under attack, and 2) immediately broadcasting "the outage is due to an attack" gives you a bit more PR cover than "umm we fucked something up badly". Companies are generally incentivized to reveal when an outage is due to something they did not directly cause.
If this is an attack, I think it's safe to say it's a failed attack.
Taking msteams offline for a day, then taking the apple app store offline for a few hours days later, doesn't hurt anyone or cause much inconvenience.
I suspect that an "Attack" would mean major systems going offline at the same time and staying down for a while.
IMHO, this is more likely a combination of:
- Large companies firing a bunch of engineers because WE HAVE AI TO REPLACE YOU.
- The remaining employees don't have the expertise to guide the AI into maintaining the systems.
- AI led by the uninformed is uninformed.
- AI tells the uninformed, "It's totally safe to reboot the gateway interface without scheduling downtime".
- Move fast and break things! (and you don't know how to fix them)
- Downtime.
What if the goal is to probe the outcome and mitigation of various outages, not to cause "The Big One" right away? If you lack the resources to effectively take down everything and combat the mitigations that would be done in response to such an attack, then knowing the fragility of the system gives you the right targeted moves to cause real damage.
Because now all DevOps teams are being forced to look at security issues that were previously back-burnered for feature work. If an attacker is probing around like this, they're likely to lose whatever backdoor they're counting on.
Noisily probing a large number of major services in a disruptive way is a great way to get all of your potential targets on high alert and patching attack options
No, just incompetent leaders at these companies. Constant layoffs, continuous turnover and offshoring.
These all have their consequences and this is what we are seeing. This is what happens when you care more about the stock price (stock manipulation via stock buybacks) rather than investing into your own company.
That's one possibility. Another hypothesis is that these distributed systems have grown so complex that some failure modes are effectively impossible to predict regardless of employee competence.
Yeah but it takes good technical leadership to stop the growth of these distributed systems beyond comprehension: the kind of leadership that awards simplicity rather than duct tape and layers of complexity for ever. Maybe not the CEO per se, but probably a VP or SVP.
Sure, but sometimes the complexity is essential, not just incidental. There are real business domains that are tremendously complex and there's no way build a simple solution without falling behind competitors. You might propose that businesses should forgo some sales to highly demanding customers in order to limit technical risk but take that too far and competitors eat your lunch.
No; you're just more focused/aware of common outages (e.g. this, the bank "outage" that wasn't, etc) that occur every week that normally resolve themselves rapidly. The FAA one is the only legitimately exceptional outage (even Azure outages are fairly run-of-the-mill).
None of the recent popular outages have been the result of a cyber attack or AI, and nobody credible is claiming otherwise.
It hasn't stopped people repeating "AI or Cyber Attack did it!" to one another in every thread, and getting more and more sure because others believe it baselessly.
That's a ridiculous statement. Outlook down for a day is not a weekly event. FAA Ground Stops aren't a weekly event.
They are things that almost never happen.
While it might be a pattern of rushed upgrades, it's still a pattern that something isn't quite right. That might just be rushed upgrades due to fear of vulnerabilities getting exploited left and right, but it's still clearly a big difference from even 6 months ago.
Then cite a source claiming the FAA, Outlook, or supposed bank outage, is linked to AI or a Cyber Attack. If you cannot link to a reliable source, it is by definition baseless.
The onus falls directly back to you to prove that these claims are unfounded. Prove that it's not a direct attack. Oh, and quoting sycophants and child rapists dont count.
The vendors themselves haven't stated such an "attack." Microsoft's released a detailed breakdown of what occurred and why on the Azure Admin portal which has been widely reposted:
So, I've cited my sources. It is now your turn to cite your "direct attack" citations. My claims have a basis, thus why I correctly called the claims of AI or Cyber Attacker baseless. That's what baseless means, as in without factual basis - which they are.
You've done nothing to explain why those are claims based in fact, and instead just deflected.
- 3 times in 25 years, but the previous failure was less than 3 years ago. Individual airlines have had their own outages at least as severe as this one numerous times in recent years
- Outlook may have never had full feature failure before, but other Microsoft services have outages all the time. Azure is a dumpster fire. GitHub even more so. Remember when Crowdstrike bricked 9 million Windows computers less than 2 years ago?
- What scale? JPMorgan Chase had an issue for a few hours. Is that unprecedented?
- NSA outages are unlikely to be leaked in the first place. we have no idea how frequently something like this occurs.
Would people even be talking about any of these if they weren’t primed to be freaking out about it? In my circle the people freaking out the most are the people most removed from tech and software development, in no small part because they have no idea what constitutes normal and are reacting to headlines and social media feeds.
The airplane grounding thing was because of a communications router in Arkansas.
When I heard the install work was done by L3 Harris, a company my department has worked with, my reaction went from "Damn, is AI sabotaging us?" to "Oh, that makes sense".
No, crazy coincidence. From an IT perspective, most of this stuff is normal stuff that can happen. The coincidence in time is interesting but not alarming to me. So many different, an unrelated, things happened with nothing tying them together (across different hardware and software).
I'm honestly wondering if our industry is fatigued from all of the activity this year. These are not examples of the best work in the industry.
Yeah I was saying this yesterday (and I’m still leaning this way) but how many more high profile outages will it take to say “hm, yeah this may not be a coincidence”.
I think my current number is sitting at about “one” at the moment.
Each of these is more or less unprecedented in severity, or at least is extremely rare. “[BIGORG] has a [near] total downtime event” 5+ times within a week isn’t coincidence; At this point the simplest explanation is that there’s a common reason.
Everything has been fine for me too on two of my three machines.
The third I just booted for the first time in a couple of months and gosh darn it my app store token expired.
Which resulted in "an unknown error has occurred" after a really long wait to re-login with apple. At my second attempt I was able to receive a two factor code on another device, but still failed to log in due to another "unknown error".
All regions. If your device isn’t seeing it you’re likely operating under the scenario op just gave. Here’s a fun set of repro steps to get your phone to vomit up the presence of a failing local loopback and what I’m fairly certain is a self-signed TLS cert.
Enable Lockdown Mode > put your phone in airplane mode > go to cellular > turn off Track Data Usage > press Reset Usage > go to on WiFi and turn it on > Scroll down to Connectivity Assist > Reset Data Usage > still in WiFi go to Edit > Scroll down to managed networks, select them and disable Connectivity Assist for any connections.
On my iPhone, from that point on, even after disabling lockdown mode, anytime I go to Safari with airplane mode on I get a giant “This Connection is Not Secure” UI. Apple hasn’t been able or willing to help since bringing this to their attention in September.
All of them? a localized outage doesn't mean shit, and AWS isn't going to advertise the team that's on call for that one service in yourtown Ohio. That's why it's cloud. The Internet has never gone down. Parts of it, sure, but just like "the banks" didn't go down, but Chase did, neither has "the Cloud". If you care about things, if shit goes down, where are your next 1000 meals coming from, how do you get there, and how do you get around any obstacles between you and them. Once you're there, how do you stop other people from taking them vs giving them away to anyone who's hungry?
My local super market had empty shelves two weeks ago because they were switching warehouses. I live in a city and we're only three + N meals away from revolution. Make your N as big as you can afford.
what a world we built in less than a generation... everything and everyone dependent on a few cloud services, we're very far from the original vision of a decentralised internet :'(
The only chance we have for a more reliable world is if these outages and hacks become more and more common. Institutions and people will be forced to wake up and find more local and diverse alternatives...
If only big tech wasn't actively hampering attempts to allow decentralisation to even function. Microsoft has _always_ abused their position of power to make it very difficult for people running their own email services to even get reliable mail delivery.
Google is just as guilty of sending correctly signed self hosted mail to the junk folder.
However they'll blatantly spew out spam to the rest of the internet, including through allowing ongoing abuse of their Google Calendar invites, Google Drive and Keep share notifications.
Good thing we threw away our one chance at anti-trust enforcement for a whole generation because eggs were expensive for a couple of months 2 years ago.
No doubt, but isn't that the point? "Customers" and "Mail Server Operators" have very different criteria for effective spam filtering!
I mean, sure, it would be nice if it were easier to send mail to gmail users from oddball domains. But that would inevitably result in gmail users getting more unwanted mail from oddball domains. Which is exactly the opposite of what they want.
What a strange concept. It kind of implies some kind of hierarchy or class system of domains, presumably with Google and Microsoft at the top.
I’ve had my own domain, hosted on Google for over ten years but when I moved it onto my own mail server Google sent all mail from me to junk. Everyone else accepted my mail fine, even Microsoft, but any Google Mail user i sent to it went to their junk folder. I had to migrate back to Google in the end.
As someone that has run mail servers for decades, don’t even get me started on all the “unwanted mail” - including malware and phishing - that I’ve had to deal with from gmail accounts.
So anything other than the major providers should be relegated to "oddball domain" and not trusted? That's the kind of centralization that's completely undermining what was supposed to be the entire point of the internet.
> So anything other than the major providers should be relegated to "oddball domain" and not trusted?
Is there a better solution you have in mind? Again, the market has spoken here. They aren't my rules and they aren't even my opinions. I ran a vanity domain on postfix and dovecot for literally decades and dealt with all of this pain.
Do we want, in the abstract, an open internet where clean communication can happen between arbitrary providers without interference from centralized authorities? Yeah, I guess.
Do customers actually care about that more than they care about spam? Duh, as it were. We're constrained to live in reality as it exists, and "accept mail from anyone with a valid SPF/DKIM/DMARC setup" is provably insufficient for keeping our inboxes clean.
I'm not sure what the better solution is, all the spam/scam emails come from the gmail.com domain so the current solution clearly ain't it. There's got to be better ways to filter emails.
"We are also seeing a lot of noise in the computer industry that the solution to reducing complexity is a simplistic approach, that is, running all software on the server, and giving users nothing but dumb terminals. It generally is not the customer's desire to throw away the billions of dollars they have invested in their current computing environment, turn all their currently productive PCs into dumb terminals and run all of their software on expensive new servers. Customers soundly rejected this model of computing when the Unix vendors tried to convince them to buy network computers."
"Putting everything on a server (and asking the world to rewrite all of its software along the way) may seem attractive if you build servers, but it ignores the real concerns and needs of many customers."
These were quotes from Paul Maritz (a VP at Microsoft) made about 27 years ago to the WSJ[1] in a back-and-forth discussion with a Sun executive (Pat Sueltz). Some of the comments were quite prescient, others aren't.
"It generally is not the customer's desire to throw away the billions of dollars they have invested in their current computing environment...."
This is true. However:
1. We now have tens of millions, maybe even hundreds of millions, of users who have little experience with traditional desktop computing. They grew up on Web services that they access primarily using smartphones, Chromebooks, or web browsers on desktop computers. There are many people with little-to-no experience with desktop applications like Microsoft Office and the Adobe Creative Suite, but they do their work on Web applications such as Office 365 and Google's suite.
2. Even for traditional desktop computer users, they often spend considerable amounts of money updating their environments once every few years. Windows has excellent backwards compatibility, but most people aren't interested in running Office 97 and Photoshop 5.5 on Windows 11 even if those programs can run on Windows 11. I like Office 97, but I need to deal with modern Office documents, plus I want the improved multi-language support and extra features that have been added in newer versions of Office. The Mac is much more aggressive about cutting off support for old programs; my PowerPC Mac OS X software can't run on my Apple Silicon Macs, and even support for Intel Mac programs will be dropped soon.
Because so many computer users these days didn't start off with traditional desktops, and also because of upgrade churn, this is how we ended up with the current situation, where many people rely on centralized solutions.
Microsoft's own email system was down recently, you may recall, so it seemed temporally pertinent; and because it's a comment on HN, not a comprehensive academic analysis of the way every big tech company has shafted decentralisation.
Decentralized Internet relies on many individuals having money to buy their own infrastructure, the expertise to maintain it and update the underlying system or hardware.
For most self hosted services, can probably get away with "set it and forget it attitude". But host a couple of publicly available services (immich to share photos with family not in your vpn for example), and asking that of the common person is a bit much.
This would also not be "free" which is pretty much the main selling point of these centralized services. Of course they use your data to create ad profiles, sell it to data brokers, and/or incorporate into data set to train LLMs.
Decentralized is possible, but it’s going up against forces that collectively have trillions of dollars at their disposal (albeit most is probably spent on stock buybacks and exec comp)
We don't need everyone to be a gardener. We just need non-chain groceries and restaurants. Folks are then free to garden and cook if they like that idea, shop and cook for a very similar cost/effort expenditure, or pay more and do none of it -- all without centralization. Well, the energy to cook is centralized in any case, other than home solar... you need to trust somebody.
> Folks are then free to garden and cook if they like that idea, shop and cook for a very similar cost/effort expenditure, or pay more and do none of it
I'm honestly not sure how that's different from the modern internet?
Right, I was making an analogy between food and the internet. Glad it fits well!
I'm seeing the ancestral comment's point that "everything and everyone [is] dependent on a few cloud services" (the chain stores/restaurants), and the followup comment's point that "[self hosting] publicly available services [...] is a bit much" (the garden), and making a point that a middle ground exists as well (the independent stores/restaurants).
You can escape the megacorp without learning to garden so long as an independent middle ground can thrive. The VPS world is alive and well, albeit less trendy than public clouds and home labs.
50 years ago we (in North America) had very centralized networks in the form of Bell/AT&T. We've reconverged to an extent, although I guess an oligopoly is better than a monopoly.
> everything and everyone dependent on a few cloud services, we're very far from the original vision of a decentralised internet :'(
A lot of the people responsible for it all going so wrong are present on this very forum. We made good money doing it, but in return what could've been one of the most positive liberating technologies in history ended up another lever of centralization and control instead.
I regret my part in it. I wonder how many feel the same. Clearly, a lot of people are still happy to push on. There's IPOs to be done and stock to vest, after all!
I remember that every time there was a choice between a positive choice and the easy choice, the average person always chose the easy one. There were plenty of alternatives to Facebook yet billions of users went to Facebook. The whole Snowden thing probably registered less acutely than the final season of Game of Thrones with the voting public.
Yes, people like us built the both good and bad platforms, but ultimately the decision which one to use was on the users, and they chose the world we have now.
Basically, I see recent history as Eternal September and the Internet turning into, through sheer volume of demand, into Super Cable TV.
It's difficult for me to understand your view so hoping you can expand on it.
> I remember that every time there was a choice between a positive choice and the easy choice, the average person always chose the easy one. There were plenty of alternatives to Facebook yet billions of users went to Facebook. The whole Snowden thing probably registered less acutely than the final season of Game of Thrones with the voting public.
I agree with this statement, but not with your conclusions from it. It is precisely because users don't care and choose the easy way out that we have a greater responsibility. Why did we fail to make the good future easy and appealing? Why did we build the bad option at all? The only honest answer I can come up with is selfishness and greed. The bad option didn't end up being easy by accident. It ended up easy because people like you and I were given mountains of money to optimize extraction and addiction.
> Yes, people like us built the both good and bad platforms, but ultimately the decision which one to use was on the users, and they chose the world we have now.
As an engineer, you are responsible for what you build and its impact on the world. If you build a very convenient, but unsafe bridge and it collapses with people on it, you don't get to just say "oh but the people chose to go on it!". We're the experts. We're the ones that are supposed to know and make good decisions, that's what we were entrusted with and that's what we failed to do.
I agree that most average people do things that lead to bad places, but I don't think it absolves us from responsibility.
Accepting your argument: let's say "we" all decided not to design the "bad" thing. But the money's still on the table for whoever does. So, "we" can all not build it, and then "they" will.
Either way it gets built, and finger wagging at "us" after the fact does nothing to solve it or prevent it happening.
I see your individualization as an attempt at bargaining for absolution through appeals to inevitability.
In reality, of course, it is a self-fulfilling prophecy.
The only way we could've prevented poor outcomes is by acting as us. As the dramatic changes of the industrial revolution have shown us, labor is immensely powerful at resisting the corrosive demands of capital if it acts in unison. Famously, the labor struggle also included dealing with scabs, or "them".
These are not new ideas. They are lessons we failed to heed out of aforementioned greed and selfishness, often dressed up as trendy tech ideology.
Efficiency and resilience are often in competition.
Price is vaguely related to efficiency, so that's what the private sector optimizes for. Think about that next time people complain about Government efficiency.
As one NASA chief said: "if we could blow up rockets as much as SpaceX did, yeah we'd be much faster and cheaper"
it's kind of a bad look for the politicians in charge when the eyes of the nation/world are on you ("you wasted how much taxpayer money on that explosion?!")
If SpaceX blows up some rockets for fast development, and that's how Elon likes it, that's how it goes.
If the NASA blows up some rockets for fast development, people get angry, politicians get scared, funding gets cut, important leaders get fired. So they have to try really hard to guarantee no failures. Everything takes way longer and costs way more. They still fail sometimes, and just as often a project is cancelled before completion, but that's way better than an explosion, politically.
I'm all for indie hosting for cultural reasons, but not in the slightest for reliability reasons.
AWS et al. (yes, even Microsoft) are far more reliable than the smaller players, let alone your own rig. The flip-side is that when we go down, we all go down together, and I'm fine with that.
I'll take my own rig 1000 times out of 1000 for reliability. This would be true even if AWS, Microsoft and the rest weren't parasites looking to constantly spy on me.
If all you need is shell access and simple tools and have modest security requirements, a few dozen Raspberry Pis (multiple models and revisions) running a stripped down immutable OS image that's only patched for severe bugs with all mutable data replicated between one another and regularly backed up to Usenet should do the trick.
Host them at home, your office, in your car, at the homes of friends and family, etc.
If all your "hosting providers" simultaneously fail catastrophically, you probably have more important things to worry about than your system.
If not, pick up a couple new Pis at retail, pop in SD cards you keep in your wallet containing the most recent base OS image, and grab a latte while your latest backup restores from Usenet over the coffee shop's free Wi-Fi.
Honestly it's surprising that this isn't the normal state of things. Knowing how brittle all this infrastructure is, I'm thankful we're giving it the amount of attention we are lately, which will likely lead to more investment in security, which should hopefully result in better availability.
I know you're speaking about the world generally, but personally I've been pulling data in locally for some years now. I have an Apple iCloud account, and Photos are in the cloud, but I am backing up to a local hard drive at least annually.
YouTube videos—download.
Music—all stuff I ripped from my own CD collections. Newer digital purchases (Bandcamp) are backed up with everything else.
Magazines from archive.org and worldradiohistory.com—local.
Early Apple history is often spiced up with tales of Woz's exploits using blue boxes, exploiting an inherently insecure centralized communications network.
Is hardware more/less/same reliable today than the old days? I’ve been in the cloud long enough that I don’t feel connected to what maintaining a machine room would be like today
It's definitely a YMMV type situation, but I would say by and large stuff used to be less reliable. That said, commoditization often leads to quality problems of its own.
I feel like this proves the opposite. Like, Apple is down, and I did not notice on any of my Apple devices, which all worked fine. iMessage worked fine. I use other providers for calendar, email, etc.
For most people in the country, despite being online constantly, they would have never known.
If anything it shows that many aspects of the Internet ARE decentralized well, even within one company. I can use a device with apps and accounts that aren’t dependent on the device manufacturer’s uptime.
Interesting timing on this, Apple down, MS down, Flight control down, NSA down, just before Trump's meeting with Xi and Putin. Attacks just before talks is typical of Putin.
I think Putin sees Trump as a tool, nothing more. He has not made his life easy (for example refusing to enter negotiations in good faith on Ukraine, backing Iran etc), if anything Trump seems beholden to Putin somehow (going by his actions like lifting sanctions or supporting russia’s demands in Ukraine), and effusive in his praise of dictators in general (Xi too).
This reminds me of the Chinese spy balloon that floated over the US. For a week or so we were getting news stories about every unidentified weather balloon or piece of airborne debris. The main connection between the incidents was that the public thought they might be connected.
Four-day NSA outage in 2000. Eventually someone leaked the news, but they tried hard to prevent that. I think leaks are more likely when people are already nervous.
The rest would absolutely not be newsworthy because they were happening before last week and not in the news. Google sucks trying to find prior art when a major event is happening so I can't find a good link but Apple specifically had an outage a few weeks/month ago and the most noise it made was Macrumors or 9to5mac
Microsoft's Exchange Online outage and the FAA outage had very big impacts, but in the case of the FAA outage wasn't even really that big of a technical issue. Those impacts rightfully were newsworthy but those two plus the PHP fire drill a few weeks ago has everyone on edge
Just because you see a bunch of murders on the news doesn't mean there's a massive increase in violent crime
> in the case of the FAA outage wasn't even really that big of a technical issue
Was the issue reliably identified then? Last I remember they swore it was definitely not the new SMART thing and failed to provide a satisfactory cause.
The latest story is summarized as a government contractor was installing a new router as part of a network upgrade, and the router generated excessive amounts of traffic that existing points in the network couldn't handle
HPE Juniper says the routers were not malfunctioning and did what they were configured to do
The implication is that the router was misconfigured in some way, but no one really knows. Given how government contractors work I buy that story, that they were doing production changes in the network at 3PM on a Friday and had a bad config
It would be really rare or impossible for a router to generate enough IP traffic on it's own (eg sourced from the router itself) to saturate either the fiber line, or the general IP processing ability of other routers.
I feel like it has to be an issue with whatever routing protocol they are using. So, low traffic in absolute (bps) terms, but something expensive for other routers to process.
I doubt that the router itself generated that traffic, I suspect either there was some kind of loop created that amplified broadcast traffic, or too many VLANs on a certain link and it forwarded more traffic than was intended, etc. There are a lot of scenarios where after a game of telephone makes it from the engineer on the ground to the Secretary of Transportation that would be close enough for him to explain it the way it was
It strains credulity that their network is designed such that a poorly inserted cable or misconfigured vlan would being the network down. If the network was that badly designed there would be constant issues.
Even more bizarrely, today it was reported that the FAA literally had someone supervising the installation, checked the work, and signed off on it, and the initial tests were perfect. So something isn’t adding up. It could be someone is pants on fire lying to save their ass, but it could also be something complicated not yet identified.
I have been explaining the FAA outage to my less technical friends this way: Think of the incident like a hyper, fast-talking New Yorker -- who just drank two pots of coffee -- walking into the quiet back room of a Southern saloon and started asking 250 questions a second at a few slow-talking good ole boys used to passing one polite sentence at a time down a single-lane dirt road. That dog just ain't gonna hunt.
Some of these outages might be successful attacks, but I'm speculating that much of the recent uptick in outages is simply the consequence of devs putting AI-written code into production without the same level of review/scrutiny that human code used to get. So, "own goals", not necessarily security incidents.
> I'm speculating that much of the recent uptick in outages is simply the consequence of devs putting AI-written code into production without the same level of review/scrutiny that human code used to get.
The code they put up more than 3-6 months ago depending on the provider, I'd agree. But 6 months ago was roughly when the best LLMs got better at spotting issues than humans, and 3 months ago was roughly when open weight models passed the same threshold.
Not that this means everyone's using the best models, of course. Number of times I see academic research and it's "we used GPT-4o and it was bad" and I check the date and oh look one was submitted yesterday: https://arxiv.org/abs/2611.20753
> If it was AWS fine, that’s not uncommon, but there have been several outages, including at the nsa, which is also unusual.
In Polish folklore there is apparently a belief that a person must be buried before Sunday (pochówek przed niedzielą?) or 'soon enough', as otherwise it will cause the deceased to pull another family member (or neighbour) into the grave with them. Perhaps the FAA and/or Microsoft outages were not fixed quickly enough?
We don’t know that outages are unusual at NSA, only that public reports of outages are unusual. It’s a large and complex enterprise, I’d wager that things break all the time.
Outages on classified networks tend to not directly impact the public.
But there was that Iranian drone mothership from China just off the east coast! The pentagon's official denial is basically an admission!
>> A US congressman Jeff Van Drew claimed he had heard from "high sources" that the flying objects were tied to Iran. [...] the New Jersey Republican said the drones were coming from an Iranian "mothership" in the Atlantic.
There are congressmen who believe space lasers change the weather, and a president who thinks COVID can be cured by injecting bleach into his arm. Being elected doesn't make you an expert in anything.
It's used to be amusing to see how much the British press misunderstands American politics and politicians, but more and more it's annoying.
Ya, being elected only means your full-time job is to represent the will of your constituents. It is the duty of every journalist to shine a light on the idiots in order to show the electorate exactly who they have elected. Covering up the crazy doesn't help.
> “A question that probably some of you are thinking of if you’re totally into that world,” Trump began, clearly thinking the question himself, “So, supposing we hit the body with a tremendous — whether it’s ultraviolet or just very powerful light — and I think you said that that hasn’t been checked, but you’re going to test it. And then I said, supposing you brought the light inside the body, which you can do either through the skin or in some other way, and I think you said you’re going to test that, too. It sounds interesting. And then I see the disinfectant, where it knocks it out in a minute. One minute. And is there a way we can do something like that, by injection inside or almost a cleaning. Because you see it gets in the lungs, and it does a tremendous number on the lungs. So it would be interesting to check that.” [emphasis mine]
He did suggest, after talking about the effectiveness of disinfects (bleach included in context) on the virus, injecting things into the body. It's very unclear, like a lot of his ramblings, what the hell he was talking about precisely, but it is reasonable to think that he wanted to inject light into the body or disinfectants or both. And if disinfectants, bleach was specifically discussed.
So did he literally say we should inject bleach into arms? No. He didn't mention where the disinfectants or light would be injected, just suggested that they could be.
He was thinking out loud about ideas that could potentially be tested, not instructing people to do anything.
There’s a reason fact checkers rated that claim as false.
It’s the kind of thing people who dislike Trump might joke about over a beer with friends. But bringing it up as a serious argument is ridiculous and intellectually dishonest.
> He was thinking out loud about ideas that could potentially be tested, not instructing people to do anything.
Neither myself nor reaperducer said he instructed anyone, you injected that into the conversation. He did say he thought it could work, literally thought we could inject bleach, other disinfectants, or light into the body to treat COVID. That is not something a sane or competent person would ever suggest, and you even agree he did suggest it.
Parent said "injecting bleach into his arm". That's completely false and misleading.
> thought we could inject bleach, other disinfectants, or light into the body to treat COVID
That's what he suggested, more or less, yes, in the context of exploring treatments. But people are so partisan and obsessed with hating Trump that they'd rather twist his words and make it sound like he was proposing to drink bleach or inject bleach in their arms. Why have nuance when you can have cool talking points.
The full transcript:
"And then I see the disinfectant, where it knocks it out in one minute. And is there a way we can do something like that, by injection inside or almost a cleaning, because you see it gets in the lungs and it does a tremendous number on the lungs, so it’d be interesting to check that, so that you’re going to have to use medical doctors with, but it sounds interesting to me. So, we’ll see, but the whole concept of the light, the way it kills it in one minute. That’s pretty powerful."
He absolutely said to inject bleach. That he didn’t specify whether to inject in in one’s arm or leg or eye or ass is irrelevant to the point being made.
Kids were posting videos of their drones on instagram and the nightly news was buying the rights...
Legitimate news organizations don't buy video from the general public. TMZ, yes. New York Times, no.
They only buy video from organizations that have been vetted and with which they have a previous relationship, like a known freelancer. They don't write checks to random kids.
I don’t know why people are downvoting you and why the parent comment is so seemingly confident that this is all just a coincidence. Like I get not giving in to conspiracy theory thinking but it’s pretty reasonable to point out that these many outages for major services one after another is pretty unusual.
Maybe this hundred-year storm happened because the gods are angry with you. Or maybe it happened because a hundred years* have passed. You'll need some evidence for that first one.
Ah, that Hundred year storm? Err, when was the last week this happened?
If these are all coincidential then we should be able to multiply the probabilities. All in all we have at least 10 different companies — including Google — with systems down. When was the last time this occurred in the same week?
1. Microsoft (Outlook, Teams, Copilot, Microsoft 365) — Nov. 12–14, 2026: A global cloud and network capacity issue disrupted email, messaging, and workplace apps for millions of users before restoration on Saturday.
2. Federal Aviation Administration (Air Traffic Control Communications & Radar) — Nov. 13, 2026: A telecommunications and router data-flooding issue triggered a multi-hour nationwide ground stop and thousands of flight delays and cancellations.
3. JPMorgan Chase (Online Banking, Transactions & ATMs) — Nov. 16, 2026: A software bug in an on-premise hardware system slowed transactions and caused login and payment issues for customers.
4. Zelle (Peer-to-Peer Payment Transfers) — Nov. 16, 2026: Users reported payment delays and errors tied to bank processing issues.
5. Wells Fargo (Online Banking & Account Access) — Nov. 16, 2026: Customers reported account access and transaction issues during the broader banking disruption.
6. Bank of America (Online Banking & Account Access) — Nov. 16, 2026: Customers flagged account access and transfer problems amid the Monday banking disruptions.
7. U.S. Bank (Online Banking & Zelle Transfers) — Nov. 16, 2026: Users reported access issues and potential delays on cross-institution Zelle payments.
8. United Airlines (Flight Departure Systems) — Nov. 16, 2026: A technical failure affecting departing flight operations caused delays for more than 90 minutes.
9. Apple (iCloud Sign-In, App Store, Apple Music, AppleCare & Apple Pay) — Nov. 17, 2026: An internal update caused outages and sign-in issues across multiple Apple services.
10. Marriott International (Marriott.com & Marriott Bonvoy App) — Nov. 17, 2026: A technical issue disrupted sign-in capabilities on the website and mobile app before resolution.
11. Zoom (Website / Optimizely Vulnerability) — Nov. 17, 2026: Affected by a security incident involving malicious code injected via the Optimizely platform across major company websites.
12. CNN (Website / Optimizely Vulnerability) — Nov. 17, 2026: Among major sites impacted by the Optimizely malicious JavaScript incident.
13. BBC (Website / Optimizely Vulnerability) — Nov. 17, 2026: Reported among major international sites affected by the Optimizely exploit.
14. Salesforce (Website / Optimizely Vulnerability) — Nov. 17, 2026: Listed among major websites impacted by the Optimizely security incident.
15. IKEA (Website / Optimizely Vulnerability) — Nov. 17, 2026: Included among major company websites affected by the Optimizely incident.
When was the last time 15 websites had issues? Oh I don't know, maybe every single time AWS has issues? Quite often. Especially if you count individual problems like the optimizely one more than once for every website that's affected, to pad your numbers and make things look worse than they are.
I think the point is that it isn’t unusual or unprecedented. It happens on a weekly basis. People start paying more attention when the attention economy finds it profitable to steer them towards it.
Aren't they unusual by definition? I just don't see how vibe-coding is a workable explanation for this particularly tight cluster of incidents, given that we've had competent coding models for at least a year now.
Low effort comment, but there are second order effects that could explain this.
Relying on code generation has pretty much been proven to atrophy skills in the short term and relying on opaque internet-connected black boxes for code generation is both leaking information and inviting back doors, should that black box be adversarial (whether due to the survival incentives of an intelligence or more likely due to the intentional biasing by the ones in control of the box).
The original poster is claiming this is a news cycle perversely creating the (perception of a) phenomenon itself, instead of the more logically-straightforward but incorrect assumption that this is a phenomenon being picked up as a news cycle.
The winner of this battle of Internet commentary will of course be the one who brings statistics and outliers to the fight. Gentlemen/women/folk: stand back to back, take ten paces and then let fly your volley of anecdotes, data, or actual reasoning, if you please.
Because a weather balloon sighting is orders of magnitude less severe and impactful on society than a global, day long office 365 outage, or a nationwide grounding during a major holiday travel window? Does it need to be explained more? The two things are not remotely comparable to make any kind of salient point. People en masse are very likely to notice once in a decade+ events occurring back to back to back without smarmily pointing to media spectacles like the weather balloon or drone incidents and saying "ah, frequency bias." In fact, such comments are even less substantive than the one you seem to be critiquing.
In fact, I find the sudden rush to explain away such events as more suspicious than the actual events themselves.
Strikes me as a little overly dismissive. The media and "average" people are always quick to jump to worst case and conspiracy level stuff for sure.
That said outlook being down for 24 hours, all flights being grounded, and apple not being able to do ANY payments would all make headlines on any given day. This isn't just "now that the populace is focused on it we'll report anything remotely related".
That said it does just seem to be a mix of bad practices catching up with bad decisions being made (hey lets update on friday..), but even I have to agree the frequency and duration is starting to look odd.
That said as with all these things, I sincerely doubt the ability of anyone to keep this shit under wraps. Only the US and its companies are being targeted AND no one from those companies screamed bloody murder about coordinated attacks/rogue AI before the gov stepped in and said "shut the fuck up."?
Maybe, but I doubt it? I know people are at risk of losing their jobs, but word gets around and I think there'd be a hell of a lot more discussion about possibilities by people who actually know what the fuck they're talking about, rather than talking heads and random nobodies who clearly don't know the industry (as often happens when any industry gets signal boosted by the media).
Hanlon's razor/Gellman amnesia still seem like the strongest answers mixed with maybe some "yeah so MS shitting the bed helped kick off a chain of tech dominoes"
Last month. 11th of October. So like, yes shit is a happening, and we know that, but that doesn't strike me as part of some massive coordinated effort going on this week, despite most of the media's implication as such.
Constructive thought exists when we have reasonable variation of perspectives. One is the typical “these events are not necessarily causally related in any sense, they’re just correlates” perspective that’s taught as a way to highlight not assuming things that look related, are related.
It’s undoubtedly a good perspective to be aware of and understand.
At the same time, we should also assume if we don’t have information to prove things aren’t correlated or reasonable evidence to think they are unrelated, we shouldn’t just dismiss them. Sometimes, correlation exists because of an underlying casual relation. It’s not always, there’s not even some frequency I could reasonably assign. So we shouldn’t dismiss it either.
I keep seeing the two polarized perspectives with few considering that it may be a mixture. Some of these events really could be related. Some may actually be pure coincidence. Some may be simply Baader–Meinhof (frequency illusion) related (we see something, therefore we look for it more, therefore we see more of it, so on) as a survival instinct.
I don’t know if there’s clarity what the current mixture is. I can read details about several of the incidents reported causes and say clearly they’re unrelated but I’m not entirely sure I trust some of the reports as it’s not in their interest to say something malicious did cause these problems.
A lot of people - particularly normies - are seeing all these infrastructure incidents happening back to back and wondering if there might be some massive coordinated attack underway. It's entirely possible. Foreign adversar(ies) pulling off something sorta like 'Stuxnet, but for everyone!' across multiple industries and tech stacks would be a pretty effective way to destabilize the US. Death by a thousand paper cuts. Our infrastructure is a lot more fragile than most people realize.
Personally I think there's a more likely explanation.
The arrival of sophisticated frontier models like Mythos is giving us an explosion in vuln discovery. CVEs, zero-days, critical infrastructure bugs, etc. There's a massive backlog of systems that suddenly need urgent attention/patching and a lot of this crap has probably been sitting out of sight and out of mind for years or more.
A tiny little security patch can cascade into major OS upgrades, TLS incompatibilities, dead protocols, dep conflicts, cert issues, and all kinds of unexpected noise. Multiply that by thousands of org (many of which are understaffed and running ancient infrastructure) and you got a recipe for chaos.
I think about the original meaning of "catfish" - long haul ocean carriers would put catfish in tanks with other fish to keep them agitated and moving so they wouldn't deteriorate during transport. AI security research is a catfish that is forcing an enormous amount of long-overdue infrastructure maintenance.
Could it be a coordinated attack? Sure. But I am leaning towards just general lack of IT competency that is getting some sunlight due to Mythos/Glasswing/etc
Why not coordinated patching. Someone dropped either a few or many vulnerabilities and all these companies need to fix it before it can be talked about.
It’s the chaotic good of conspiracy theories we can use to imagine a world where there’s a plan and things are going according to it vs rudderless happenstance.
If you were a coordinated attacker you would probably try to make all the outages happen simultaneously since the effect would be compounded. Having each outage happen one after another is kind of a best case scenario. The fact that the outages have been recoverable so far is heartening. Finally, so far no evidence of data corruption or exfiltration. If you're a foreign adversary, downtime is nice but its a pretty weak attack. All this being said, don't be surprised if an openly foreign adversary like Iran tries to take credit. It costs them little and the way their regime is set up being able to demonstrate capacity to degrade the US helps them domestically. Whether it would be true or not is unrelated.
I guess it depends on the goal, right? If your goal is to create headlines for X full days through attacks (say 10 days) while burning the fewest zero days, to prove to an adversary you can do a lot of damage as leverage to get them to accede to your demands, it looks like what we’re seeing. Note Putin, Xi, Trump meeting is next week.
I don’t want to dramatize this, but one thing is certain: we depend too much on big tech, and some of those dependencies are crucial to our daily lives.
It reminds me of the guy who was banned by Apple for the wrong reason [1]. It took him several days, and he had to reach important people at Apple, to get his account back.
That was quite the cautionary tale to not put all your eggs in one basket.
Apple certainly makes that hard to do, especially if you publish any apps on their devices. Or use any of their services that have tight integration in their ecosystem.
Interesting that HomeKit makes the general status page and not the developer one. Makes me wonder how they define each entry, and what makes up a system in one vs. the other.
My guess is that it is something along the lines of "Would a normal user be affected by it".
A normal user might expect to access their home devices from outside the house (which is what I presume "HomeKit" covers). But only a "publisher" (for want of a better term) would need to access App Store Connect, or News Publisher or similar.
Maybe a better way to think about recent outages is “increased chaos in the system”.
AI race redirected resources and attention away from operational concerns. There’s also seasonal increase in traffic in many industries, which may put extra pressure to deploy updates faster.
Maybe an AWS thing? as i recall a major chunk of icloud data stores is S3 based and they also use a bit of AWS computeoutside of their own data centers.
Well, I'm glad I'm running Linux on all of my personal computers and self-host files, photos, music and movies. It's ridiculous that in order to use my own computer I'd have to log into an account at the company in the US where I don't even live.
I had an interesting experience picking up a rental car late at night from an airport recently. There was a long line at the counter. After a while gossip reached back to my position that all their computer systems were down. I thought this seemed strange, perhaps their internet connection was down I thought. So I pulled out my phone and tried to load the company's web site. Definitely down. Anyway the interesting thing is they managed to rent cars to 20 people nevertheless. They took photocopies of licenses and credit cards. They filled out paper forms (with carbon copies). Quite amazing. Some of the staff were of an age where they had probably worked the manual process every day, but still.
I've seen that in other industries, too. A few years back one of the major providers for car dealers (maybe -the- largest provider) went down for a couple days as I recall, and it was interesting to see the subreddit for service advisors discuss bringing out all the old paperwork and doing everything manually. Same for sales. Turns out that when real money is on the line, they can figure out alternatives pretty fast.
A smaller anecdote, but I ordered a HDD that was DOA from a semi-large national electronics retailer.
They were having IT issues, so they just wrote something down on a piece of paper, went to grab something from the warehouse, noted the SKU/serial number, and gave it to me.
Is it just me or is technology letting us down. Like suddenly AI is scaling the ability to poke holes way way way faster than we can slap AI duct tape over said holes.
My theory with all of this is that AI vulnerability detection has risen significantly, as well as AI tools that can exploit them such as ARTEX (which the FBI has put a briefing out for).
These interruptions are potentially a combination of companies racing to patch faster to keep up, or a large exploit has been found, or importantly AI exploit tools like ARTEX are being used in the wild.
Also just non-malicious agents haphazardly running amok, loading web pages, calling APIs, releasing app updates, etc. The amount of stuff happening digitally has 10000x’d in a year.
All these major companies and services going down one after another over the past few days feels almost like a disaster preparedness drill .... the kind that military conducts when anticipating something catastrophic.
On a somewhat related note, Safeway's product ordering search went down today, showing nothing but a spinning icon. If I asked for a category, such as frozen food, I'd get 0 entries returned. The only groceries I could order were ones I had ordered before, because I could see that list. There were also login problems. I did eventually get an order in, which was delivered promptly.
That happened to me regularly when I would use one of their other brands (Tom Thumb and/or Albertsons) sites as well. This isn’t a new thing for them at all. I would usually fix it by switching to Kroger.
Whatever vets Apple Pay applications has been broken for months IME. When I finally got someone to respond and fix things, it turns out they deleted our entire application, so now I have to redo it all.
Frankly unsure if Apple Pay is worth the bother for us, at this point.
Assume most software and IOT devices are all exploitable by constantly advancing AI. Then assume this leads to an arms race between attackers and defenders. Nation state level threat actors then begin to target each others systems, in addition to all sorts of other hacker groups, lone wolves, etc.
The number of vulernable systems far outweighs all of the patching being done by the defenders. Exponentially increasing hacks and unexplained outages begin occurring in many systems.
AI patching turns out to not only cause the initial downtime from the accrued technical debt, in many situations the rush to put out a patch for a known exploit itself causes more outages because AI is not advanced enough to have enough context for every system it is trying to fix.
But so many of these systems can't be taken offline because the modern world depends on them. The only solution is going to be to destroy the open internet, strict firewalls between countries and regions, and excessive internal surveillance to track down any domestic threats.
The open/global internet will have to get locked down.
Aah, my favorite gaslighting page that loves to show all green while I’m trying to get to a service that returns 500.
If it’s actually red, that means things are so bad Apple can’t mask it anymore.
(btw, there was a post going around guessing that LLMs has discovered 0day with exploits in the wild so serious, that several huge companies took their stuff offline to fix it)
frenchie4111 | a day ago
I believe Apple is hosted (almost) entirely on their own hardware? So there's no reason an outage would be correlated with other outages
edit: Their status page is showing partial outage https://www.apple.com/ca/support/systemstatus/
[OP] willy_k | a day ago
https://substack.com/@thedreydossier/note/c-389611426
Source: NSA Internal Network Down https://news.ycombinator.com/item?id=50534145
wat10000 | a day ago
[OP] willy_k | a day ago
Separately: People conspire. Take whatever level of manager/executive etc. meeting you’re privy to, the fudging around external transparency / comms that goes on there, and scale that out to a country. Now factor in current admin.
wat10000 | 21 hours ago
Despite the name, "conspiracy" as in "conspiracy theory" isn't about the conspiracy per se. It's about the sort of handwavey, vibes-based thinking used in certain common arguments that also involve conspiracies. The issue I have with this claim isn't the conspiracy part, it's the complete lack of any actual explanation or any reason to believe these things are connected beyond the timing and some vague feelings. Those just happen to be commonly found in conspiracy-based claims such as this one.
vjulian | 23 hours ago
As for conspiratorial, many explanations fit the (highly limited in circulation) facts. There is no basis for demoting any explanation as ‘conspiratorial’.
wat10000 | 22 hours ago
I label it as conspiratorial because there's no explanation given for why there would be a connection, beyond the timing and some vague handwaving about a government project with an interesting name.
Also the linked Substack is very, very obviously a conspiracy theory nut. I instantly got that vibe clicking through the provided link, and poking around reinforces that strongly. For example, in https://substack.com/home/post/p-197154724 she draws a line between Genesis Mission and an electronic health records program called MHS Genesis on the basis of nothing but the name. And the supposed connection is that Genesis Mission will train AI on the data from MHS Genesis. Which makes very little sense. How useful would electronic health records be as training data? This question is not explored at all. It's just taken as a given that this would be a useful thing for the government. But that's just the beginning. She then pivots to saying that this massive supercomputer "where the body becomes the battlefield" is being built under the demolished East Wing of the White House. That's lip-strummingly crazy. That's "Tesla was killed because he invented free energy at Wardenclyffe" level nonsense. That's the sort of thing that acts as evidence against any claim it's near.
lowdownbutter | 22 hours ago
tedd4u | 23 hours ago
mynameisvlad | a day ago
JoeBOFH | a day ago
owenthejumper | a day ago
eyeundersand | a day ago
eyeundersand | a day ago
alt227 | a day ago
Apple are no different to any other service provider.
alt227 | a day ago
No and it never has been. Apple is one of AWS biggest customers.
“Apple’s cloud business is hugely dependent on Amazon The iPhone maker pays $30 million a month to use AWS, CNBC reports”
https://www.educative.io/blog/does-apple-use-aws
https://123heerasingh.medium.com/apple-icloud-dependability-...
EDIT: There is a really good breakdown of it here:
https://www.awscertificationhandbook.com/blog/does-apple-use...
dilyevsky | 23 hours ago
radicaldreamer | 23 hours ago
dilyevsky | 23 hours ago
alt227 | 22 hours ago
wrxd | 23 hours ago
Apple services mostly run in Apple's own DC. Storage make extensive use of public cloud providers
tomrod | 23 hours ago
tims33 | 23 hours ago
ThePowerOfFuet | 20 hours ago
>When exploring the question does Apple use AWS, it is useful to examine the specific services that benefit from cloud infrastructure.
larrywright | a day ago
paulhebert | 23 hours ago
We were testing this and it suddenly stopped working. (Both the new undeployed version but also the production version which hasn’t changed in weeks.) Seems connected to the outage.
Forgeties79 | 23 hours ago
cjbconnor | a day ago
itopaloglu83 | a day ago
rickreynoldssf | 23 hours ago
lrvick | 23 hours ago
Apple has literally no idea what they are doing with Linux and cloud.
Symbiote | 23 hours ago
mingus88 | 23 hours ago
They also use AWS and run servers in China using whatever cloud provider the Party uses
Like you’d expect a trillion dollar global tech enterprise to be, they run a ton of stuff geared toward whatever meets their business objectives
ibejoeb | 23 hours ago
rdtsc | 22 hours ago
prmoustache | 22 hours ago
Nevermark | 18 hours ago
Yeah.
lrvick | 17 hours ago
Such tactics mean a single compromised engineer laptop means a compromise of production, and thus a data breach is all but inevitable.
I would disqualify anyone seriously suggesting Salt, Puppet, Ansible or otherwise advocating for live-mutating production servers in a sysadmin interview today.
solyin | 15 hours ago
lrvick | an hour ago
Custom operating systems are not scary. Make a UKI so your entire OS is just a kernel, init, and a target daemon payload. Treat servers like firmware. The less code in memory the better.
That said, I am working on an general purpose hardened server OS to be released soon publicly covered in my last talk https://lance.dev/talks/bsm2026
edf13 | 11 hours ago
lrvick | an hour ago
prmoustache | 11 hours ago
> Such tactics mean a single compromised engineer laptop means a compromise of production, and thus a data breach is all but inevitable.
I hope you don't believe that an immutable image cannot be compromised. The commits done to the build system come from the same engineer laptops and, in a world where CVE are published ao frequently, said compromised image could be deployed largely in only a few minutes hours.
Also there is something to be said with lack of debugging tools. While I agree that reducing the surface of attack is the way to go, I am pretty sure the reason it takes so long for some of the big infra outages we have experienced from various cloud vendors comes from the inability of engineers to log on some systems, observe and potentially fix temporarily by hand to stop a race condition. There is often a race condition and serpent eating its own tail situation involved and logs might not even reach the monitoring infra anymore.
Going back to manage pets is not the way to go but I am not entirely certain removing ability for a subset of engineers to log on systems is exactly the way to go.
lrvick | an hour ago
Obviously, but I will trust a unikernel with just kernel/init and an unprivileged sandboxed daemon to have dramatically less attack surface than a redhat system. Reducing attack surface is the job.
> The commits done to the build system come from the same engineer laptops and, in a world where CVE are published ao frequently, said compromised image could be deployed largely in only a few minutes hours.
That is what signed commits and signed reviews are for, and a CI that mandates both. I mentioned quorum control and this is the method I always use. I founded the only Linux distribution that does this.
> Also there is something to be said with lack of debugging tools.
In some cases one may wish to expose a custom, unprivileged, and sandboxed ssh daemon that provides basic logging and read only diagnostic tools to admins. This is usually fine. Hardening and observability can co exist. No need for a real ssh and huge traditional desktop linux shell interface attack surface.
But if I want to do major debugging with dangerous tools in prod, which should almost never be required, it should have to be with consent of another team member to add a -new- image to the fleet with those debug tools, temporarily.
This would only be required when the only way to debug is with production secrets, but the default should always be to reproduce on a duplicate system with dummy secrets.
aprdm | 20 hours ago
electric_mayhem | 15 hours ago
They don't.
swiftcoder | 23 hours ago
That's not the entirely of their cloud exposure, however. A ton of their stuff runs in AWS, I wouldn't be surprised if they had things on other cloud providers as well.
mixdup | 22 hours ago
gr33nq | 21 hours ago
blinding-streak | 22 hours ago
etc-hosts | 20 hours ago
icecheese | 23 hours ago
redwood | 23 hours ago
tsimionescu | 22 hours ago
SOLAR_FIELDS | 17 hours ago
basch | 23 hours ago
TSA is down too.
MiroslavPokorny | 14 hours ago
Do you really think that when these events happen, every single computer is patched at the SAME time ?
Does that sound intelligent ?
Does this happen in other industries ? Does the electric company shutdown every single transformer/station/big machine i dont know the name and perform servicing at the SAME time ?
Of course not, if they do they shouldnt.
chickensong | 9 hours ago
MiroslavPokorny | 9 hours ago
Bluescreenbuddy | 23 hours ago
NichoPaolucci | 20 hours ago
Apple was fine, but this poor old woman was at the genius bar trying to get some accounts connected. The genius at the Apple store was trying to explain the outage - and it just was NOT clicking for the woman. He must have said 4 variations of "Yes your accounts SHOULD connect, but Microsoft is having a major outage right now so they do not."
lazyasciiart | 14 hours ago
N19PEDL2 | 23 hours ago
Bricked from the very beginning.
garrettgarcia | 21 hours ago
SilentM68 | 20 hours ago
apparent | 19 hours ago
inshard | a day ago
zzzeek | a day ago
swasheck | a day ago
https://www.apple.com/support/systemstatus/
Bruises | 23 hours ago
Both the app store and dropbox (??) have been unable to connect for the last six hours for me while other apps on ios are fine. Dropbox's status website says everything is fine, and I can't understand how that could be affected by apple services being down.
Edit: Dropbox finally caught up and added "Sign in with Apple is not working as expected for some users" to their status page
adamsiem | a day ago
My calendar failed to login. Then I couldn't login to account.apple.com on any device.
Someone1234 | 23 hours ago
rconti | a day ago
kiddz | a day ago
wankerrific | a day ago
monkpit | a day ago
jarhed | a day ago
SecretDreams | a day ago
uludag | a day ago
Anything that goes fast, accelerates, enhances, boosts, dazzles is, in some way, whether you realize it or not, whether you agree or not, is connected to AI.
Anything faulty, slow, outdated, error-prone, is clearly due to some human in some way shape or form. If there's an ostensible cause from AI, the actual problem is probably from a human not reviewing the code or following processes, or writing a bad prompt.
NoGravitas | 23 hours ago
DonHopkins | 23 hours ago
lynndotpy | a day ago
I started getting notifications that my Apple ID was logged out of iCloud, and I am not suffering for it, since I am not using iCloud for much.
corps_and_code | a day ago
kridsdale3 | 23 hours ago
detourdog | 22 hours ago
ruther4d | a day ago
xyst | 23 hours ago
On another note, using non-iCloud accounts with those apps often results in a “degraded" experience. Many of the features offered with iCloud accounts on those apps are missing when connected to a different service.
kridsdale3 | 23 hours ago
Them missing features you refer to are not some kind of anticompetetive move, it's that protocols don't advance to support these things.
ksajadi | a day ago
MBCook | a day ago
andrewinardeer | a day ago
jeffnv | a day ago
AndrewOMartin | 23 hours ago
seatac76 | 23 hours ago
shikshake | 23 hours ago
interestica | 23 hours ago
asimovDev | 10 hours ago
rickreynoldssf | 23 hours ago
I don't doubt for a second that the same kinds of morons are prepping the same attack for Thursday.
MBCook | 18 hours ago
gizajob | 23 hours ago
random__duck | 23 hours ago
odiroot | 23 hours ago
HarHarVeryFunny | 23 hours ago
Oras | 23 hours ago
Carrok | 23 hours ago
bita_nidir | 21 hours ago
SerCe | a day ago
ryanmcbride | a day ago
orblivion | a day ago
kokanee | a day ago
nomilk | 23 hours ago
Meanwhile, Apple's status pages list Apple Pay as green: https://www.apple.com/support/systemstatus/
ruther4d | a day ago
Edit: and apparently Apple takes appointments but has no methodology to cancel them due to catastrophic outages like this, nor do they have any ability to conduct non-digital transactions…typical tech
mitchweaver | 23 hours ago
I can't imagine it's gotten any better.
01284a7e | 23 hours ago
BobBagwill | 23 hours ago
jazzyjackson | 22 hours ago
reaperducer | 22 hours ago
I see what you did there.
ditn | 23 hours ago
jjtheblunt | 22 hours ago
ditn | 22 hours ago
jjtheblunt | 21 hours ago
linhns | 18 hours ago
jjtheblunt | 18 hours ago
sl0wbrainRot | 23 hours ago
The scale of the data center build outs are not technically necessary. It's all designed to make big tech too big to fail, scare aging pols deeply invested in big tech stocks; the real infinite money glitch...scare the bejeezus out of the aging political class and they'll bankrupt the world for you
The only SaaS app I really need is a Dropbox for offsite backup.
If shit goes to shit where we are people can ping remote contacts direct to fan out the news; proper point to point topology.
Instead we are forced into a hub and spoke topology and we are not the hubs
All other backend infrastructure could be whittled down to fiber and wireless network gear where upon inspection all ya can see is electrons or scrambled packets.
LPisGood | a day ago
orblivion | a day ago
I think such things are happening more and more year after year, which is newsworthy in and of itself. I'm not denying that. I'm not even denying that AI/cyber incidents on a big scale haven't happened and aren't coming. This is all newsworthy. I just have doubts about how people are interpreting the events of this particular week.
thesuitonym | a day ago
SoftTalker | 22 hours ago
neom | a day ago
mlinsey | 23 hours ago
busymom0 | 21 hours ago
https://old.reddit.com/r/iOSProgramming/comments/1xypp96/app...
caycep | a day ago
The_Blade | 23 hours ago
not just software is getting enshittified, people. they are coming after your Snickers, and your Hersheys, and your Reese's
elevation | 23 hours ago
Honestly after a childhood of cheap knock-off chocolate, the real stuff tastes weird, and the cheap stuff remains a comfort food.
UltraSane | 23 hours ago
rc5150 | 21 hours ago
dopa42365 | 21 hours ago
isodev | a day ago
nomilk | a day ago
nantes | a day ago
dkdbejwi383 | 23 hours ago
orblivion | 23 hours ago
ghastmaster | 23 hours ago
aktuel | 23 hours ago
deathanatos | 23 hours ago
dkdbejwi383 | 22 hours ago
watson | 23 hours ago
DonHopkins | 23 hours ago
reaperducer | 22 hours ago
FTFY
mentalgear | 23 hours ago
Forgeties79 | 23 hours ago
seizethecheese | 23 hours ago
It's particularly egregious in cases like this. For example, there were so many "scary virus" stories post-covid that never would have gotten press.
chrisbrandow | 20 hours ago
testfrequency | a day ago
For context, the Retail stores are also unable to do transactions, process repairs, or use any of their internal services for customers. They are essentially fully offline.
xyst | 23 hours ago
Apple is just yet another company that doesn’t give a shit
alexgoodhart | 18 hours ago
We move onward to improvisatory business practice. Cyber should take a lesson.
leca | a day ago
sperr11 | a day ago
georgeburdell | 22 hours ago
andrewinardeer | a day ago
mitchweaver | 23 hours ago
Real big brains at Apple.
123123321 | a day ago
123123321 | a day ago
dude250711 | a day ago
worik | a day ago
The very oldest TCP/IP gets on trucking, or we would not even know
hippycruncher22 | a day ago
longtimelistnr | a day ago
lifeisgood99 | 23 hours ago
swat535 | 23 hours ago
Anyone claiming it's a coincident or that it's a guaranteed cyberattack is speculating.
All we can do right now is to observe and note the patterns.
meowface | 21 hours ago
One reason is that if a company is under attack, 1) usually, you can tell you're under attack, and 2) immediately broadcasting "the outage is due to an attack" gives you a bit more PR cover than "umm we fucked something up badly". Companies are generally incentivized to reveal when an outage is due to something they did not directly cause.
jm4 | 23 hours ago
mmh0000 | 23 hours ago
Taking msteams offline for a day, then taking the apple app store offline for a few hours days later, doesn't hurt anyone or cause much inconvenience.
I suspect that an "Attack" would mean major systems going offline at the same time and staying down for a while.
IMHO, this is more likely a combination of:
tedd4u | 23 hours ago
goda90 | 23 hours ago
mmh0000 | 22 hours ago
Because now all DevOps teams are being forced to look at security issues that were previously back-burnered for feature work. If an attacker is probing around like this, they're likely to lose whatever backdoor they're counting on.
NeutralCrane | 19 hours ago
xyst | 23 hours ago
These all have their consequences and this is what we are seeing. This is what happens when you care more about the stock price (stock manipulation via stock buybacks) rather than investing into your own company.
nradov | 23 hours ago
kccqzy | 22 hours ago
nradov | 22 hours ago
DonHopkins | 23 hours ago
Someone1234 | 23 hours ago
None of the recent popular outages have been the result of a cyber attack or AI, and nobody credible is claiming otherwise.
It hasn't stopped people repeating "AI or Cyber Attack did it!" to one another in every thread, and getting more and more sure because others believe it baselessly.
eterm | 23 hours ago
That's a ridiculous statement. Outlook down for a day is not a weekly event. FAA Ground Stops aren't a weekly event.
They are things that almost never happen.
While it might be a pattern of rushed upgrades, it's still a pattern that something isn't quite right. That might just be rushed upgrades due to fear of vulnerabilities getting exploited left and right, but it's still clearly a big difference from even 6 months ago.
TaupeRanger | 23 hours ago
- FAA grounding: 3 times in 25 years
- Outlook/MS full day outage: full failure never happened before for this length
- Multi-institution bank outages: rare on this scale
- NSA major internal network outage: also rare
- Apple login/multi-service outages: couple times per year
Someone1234 | 22 hours ago
Then cite a source claiming the FAA, Outlook, or supposed bank outage, is linked to AI or a Cyber Attack. If you cannot link to a reliable source, it is by definition baseless.
So by all means "base" it. Cite facts.
rc5150 | 21 hours ago
Someone1234 | 21 hours ago
https://www.reddit.com/r/sysadmin/comments/1xxiasy/microsoft...
The FAA also released a statement detailing new router installation related problems:
https://www.cnbc.com/2026/11/17/faa-ground-stop-routers.html
Bank of America cited a technical fault:
https://www.cbsnews.com/news/banks-online-service-outages-jp...
So, I've cited my sources. It is now your turn to cite your "direct attack" citations. My claims have a basis, thus why I correctly called the claims of AI or Cyber Attacker baseless. That's what baseless means, as in without factual basis - which they are.
You've done nothing to explain why those are claims based in fact, and instead just deflected.
linhns | 18 hours ago
NeutralCrane | 20 hours ago
- Outlook may have never had full feature failure before, but other Microsoft services have outages all the time. Azure is a dumpster fire. GitHub even more so. Remember when Crowdstrike bricked 9 million Windows computers less than 2 years ago?
- What scale? JPMorgan Chase had an issue for a few hours. Is that unprecedented?
- NSA outages are unlikely to be leaked in the first place. we have no idea how frequently something like this occurs.
Would people even be talking about any of these if they weren’t primed to be freaking out about it? In my circle the people freaking out the most are the people most removed from tech and software development, in no small part because they have no idea what constitutes normal and are reacting to headlines and social media feeds.
WarmWash | 23 hours ago
When I heard the install work was done by L3 Harris, a company my department has worked with, my reaction went from "Damn, is AI sabotaging us?" to "Oh, that makes sense".
L3 has been on the rocks for the last few years.
octoberfranklin | 17 hours ago
m4r1k | a day ago
* FAA
* Banks
* Optimizely malware
* nsanet
* Apple
Time to panic?
annoyingnoob | 23 hours ago
I'm honestly wondering if our industry is fatigued from all of the activity this year. These are not examples of the best work in the industry.
soupfordummies | 23 hours ago
I think my current number is sitting at about “one” at the moment.
[OP] willy_k | 23 hours ago
david_shi | 9 hours ago
The question is whether the problems all just around patching or if there are underreported exploits that are happening.
jrflo | 23 hours ago
tedd4u | a day ago
prometheus1992 | 23 hours ago
HFguy | 23 hours ago
tedd4u | 23 hours ago
za_creature | 23 hours ago
peanut-walrus | 22 hours ago
za_creature | 17 hours ago
The third I just booted for the first time in a couple of months and gosh darn it my app store token expired.
Which resulted in "an unknown error has occurred" after a really long wait to re-login with apple. At my second attempt I was able to receive a two factor code on another device, but still failed to log in due to another "unknown error".
cududa | 23 hours ago
Enable Lockdown Mode > put your phone in airplane mode > go to cellular > turn off Track Data Usage > press Reset Usage > go to on WiFi and turn it on > Scroll down to Connectivity Assist > Reset Data Usage > still in WiFi go to Edit > Scroll down to managed networks, select them and disable Connectivity Assist for any connections.
On my iPhone, from that point on, even after disabling lockdown mode, anytime I go to Safari with airplane mode on I get a giant “This Connection is Not Secure” UI. Apple hasn’t been able or willing to help since bringing this to their attention in September.
rafram | 23 hours ago
andreigheorghe | 23 hours ago
Exoristos | 23 hours ago
SoftTalker | 22 hours ago
MiroslavPokorny | 14 hours ago
Good to be at the arse end of the world, nobody cares about us here. Im going to guess in the future this distance will be an even bigger asset.
PowerElectronix | a day ago
joanwestenberg | a day ago
rebello95 | 23 hours ago
comrade1234 | 23 hours ago
j45 | 23 hours ago
Did anyone experience an odd momentary issue with Apple Pay yesterday?
tedd4u | 23 hours ago
Following a string of high-profile online service outages this past week, several Apple services appear to be experiencing issues today.
https://tech.yahoo.com/general/breaking-news/article/apple-c...
xyst | 23 hours ago
solyin | 22 hours ago
My local super market had empty shelves two weeks ago because they were switching warehouses. I live in a city and we're only three + N meals away from revolution. Make your N as big as you can afford.
DonHopkins | 23 hours ago
https://www.youtube.com/watch?v=rSjK2Oqrgic
whh | 23 hours ago
blop | 23 hours ago
The only chance we have for a more reliable world is if these outages and hacks become more and more common. Institutions and people will be forced to wake up and find more local and diverse alternatives...
lukeify | 23 hours ago
cube00 | 23 hours ago
However they'll blatantly spew out spam to the rest of the internet, including through allowing ongoing abuse of their Google Calendar invites, Google Drive and Keep share notifications.
coldpie | 23 hours ago
RunSet | 15 hours ago
https://www.politico.com/news/magazine/2026/10/18/egg-prices...
syshyk | 22 hours ago
StilesCrisis | 22 hours ago
harry8 | 20 hours ago
StilesCrisis | 19 hours ago
davkan | 19 hours ago
gimlichgu | 18 hours ago
ajross | 18 hours ago
I mean, sure, it would be nice if it were easier to send mail to gmail users from oddball domains. But that would inevitably result in gmail users getting more unwanted mail from oddball domains. Which is exactly the opposite of what they want.
funnybeam | 11 hours ago
What a strange concept. It kind of implies some kind of hierarchy or class system of domains, presumably with Google and Microsoft at the top.
I’ve had my own domain, hosted on Google for over ten years but when I moved it onto my own mail server Google sent all mail from me to junk. Everyone else accepted my mail fine, even Microsoft, but any Google Mail user i sent to it went to their junk folder. I had to migrate back to Google in the end.
As someone that has run mail servers for decades, don’t even get me started on all the “unwanted mail” - including malware and phishing - that I’ve had to deal with from gmail accounts.
Hugsbox | 7 hours ago
ajross | 3 hours ago
Is there a better solution you have in mind? Again, the market has spoken here. They aren't my rules and they aren't even my opinions. I ran a vanity domain on postfix and dovecot for literally decades and dealt with all of this pain.
Do we want, in the abstract, an open internet where clean communication can happen between arbitrary providers without interference from centralized authorities? Yeah, I guess.
Do customers actually care about that more than they care about spam? Duh, as it were. We're constrained to live in reality as it exists, and "accept mail from anyone with a valid SPF/DKIM/DMARC setup" is provably insufficient for keeping our inboxes clean.
Hugsbox | 20 minutes ago
MiroslavPokorny | 14 hours ago
The-Bus | 22 hours ago
"Putting everything on a server (and asking the world to rewrite all of its software along the way) may seem attractive if you build servers, but it ignores the real concerns and needs of many customers."
These were quotes from Paul Maritz (a VP at Microsoft) made about 27 years ago to the WSJ[1] in a back-and-forth discussion with a Sun executive (Pat Sueltz). Some of the comments were quite prescient, others aren't.
1 = https://www.wsj.com/articles/SB942104638509774099
linguae | 17 hours ago
This is true. However:
1. We now have tens of millions, maybe even hundreds of millions, of users who have little experience with traditional desktop computing. They grew up on Web services that they access primarily using smartphones, Chromebooks, or web browsers on desktop computers. There are many people with little-to-no experience with desktop applications like Microsoft Office and the Adobe Creative Suite, but they do their work on Web applications such as Office 365 and Google's suite.
2. Even for traditional desktop computer users, they often spend considerable amounts of money updating their environments once every few years. Windows has excellent backwards compatibility, but most people aren't interested in running Office 97 and Photoshop 5.5 on Windows 11 even if those programs can run on Windows 11. I like Office 97, but I need to deal with modern Office documents, plus I want the improved multi-language support and extra features that have been added in newer versions of Office. The Mac is much more aggressive about cutting off support for old programs; my PowerPC Mac OS X software can't run on my Apple Silicon Macs, and even support for Intel Mac programs will be dropped soon.
Because so many computer users these days didn't start off with traditional desktops, and also because of upgrade churn, this is how we ended up with the current situation, where many people rely on centralized solutions.
bmitc | 17 hours ago
lukeify | 16 hours ago
bmitc | 15 hours ago
xyst | 23 hours ago
For most self hosted services, can probably get away with "set it and forget it attitude". But host a couple of publicly available services (immich to share photos with family not in your vpn for example), and asking that of the common person is a bit much.
This would also not be "free" which is pretty much the main selling point of these centralized services. Of course they use your data to create ad profiles, sell it to data brokers, and/or incorporate into data set to train LLMs.
Decentralized is possible, but it’s going up against forces that collectively have trillions of dollars at their disposal (albeit most is probably spent on stock buybacks and exec comp)
WarmWash | 23 hours ago
syshyk | 22 hours ago
"If it goes wrong, they fix it".
hunter2_ | 23 hours ago
We don't need everyone to be a gardener. We just need non-chain groceries and restaurants. Folks are then free to garden and cook if they like that idea, shop and cook for a very similar cost/effort expenditure, or pay more and do none of it -- all without centralization. Well, the energy to cook is centralized in any case, other than home solar... you need to trust somebody.
William_BB | 22 hours ago
I'm honestly not sure how that's different from the modern internet?
hunter2_ | 14 hours ago
I'm seeing the ancestral comment's point that "everything and everyone [is] dependent on a few cloud services" (the chain stores/restaurants), and the followup comment's point that "[self hosting] publicly available services [...] is a bit much" (the garden), and making a point that a middle ground exists as well (the independent stores/restaurants).
You can escape the megacorp without learning to garden so long as an independent middle ground can thrive. The VPS world is alive and well, albeit less trendy than public clouds and home labs.
jakeinspace | 23 hours ago
vrganj | 23 hours ago
A lot of the people responsible for it all going so wrong are present on this very forum. We made good money doing it, but in return what could've been one of the most positive liberating technologies in history ended up another lever of centralization and control instead.
I regret my part in it. I wonder how many feel the same. Clearly, a lot of people are still happy to push on. There's IPOs to be done and stock to vest, after all!
throwayay58373 | 22 hours ago
I remember that every time there was a choice between a positive choice and the easy choice, the average person always chose the easy one. There were plenty of alternatives to Facebook yet billions of users went to Facebook. The whole Snowden thing probably registered less acutely than the final season of Game of Thrones with the voting public.
Yes, people like us built the both good and bad platforms, but ultimately the decision which one to use was on the users, and they chose the world we have now.
Basically, I see recent history as Eternal September and the Internet turning into, through sheer volume of demand, into Super Cable TV.
It's difficult for me to understand your view so hoping you can expand on it.
vrganj | 21 hours ago
I agree with this statement, but not with your conclusions from it. It is precisely because users don't care and choose the easy way out that we have a greater responsibility. Why did we fail to make the good future easy and appealing? Why did we build the bad option at all? The only honest answer I can come up with is selfishness and greed. The bad option didn't end up being easy by accident. It ended up easy because people like you and I were given mountains of money to optimize extraction and addiction.
> Yes, people like us built the both good and bad platforms, but ultimately the decision which one to use was on the users, and they chose the world we have now.
As an engineer, you are responsible for what you build and its impact on the world. If you build a very convenient, but unsafe bridge and it collapses with people on it, you don't get to just say "oh but the people chose to go on it!". We're the experts. We're the ones that are supposed to know and make good decisions, that's what we were entrusted with and that's what we failed to do.
I agree that most average people do things that lead to bad places, but I don't think it absolves us from responsibility.
buggyipadmettoo | 17 hours ago
Accepting your argument: let's say "we" all decided not to design the "bad" thing. But the money's still on the table for whoever does. So, "we" can all not build it, and then "they" will.
Either way it gets built, and finger wagging at "us" after the fact does nothing to solve it or prevent it happening.
funnybeam | 11 hours ago
That is a choice you made and you are responsible.
(using “you” figuratively here, not accusing you personally of anything)
vrganj | 10 hours ago
In reality, of course, it is a self-fulfilling prophecy.
The only way we could've prevented poor outcomes is by acting as us. As the dramatic changes of the industrial revolution have shown us, labor is immensely powerful at resisting the corrosive demands of capital if it acts in unison. Famously, the labor struggle also included dealing with scabs, or "them".
These are not new ideas. They are lessons we failed to heed out of aforementioned greed and selfishness, often dressed up as trendy tech ideology.
BeFlatXIII | 5 hours ago
AceJohnny2 | 23 hours ago
Price is vaguely related to efficiency, so that's what the private sector optimizes for. Think about that next time people complain about Government efficiency.
As one NASA chief said: "if we could blow up rockets as much as SpaceX did, yeah we'd be much faster and cheaper"
ttoinou | 18 hours ago
makeitdouble | 18 hours ago
AceJohnny2 | 17 hours ago
ploxiln | 14 hours ago
If the NASA blows up some rockets for fast development, people get angry, politicians get scared, funding gets cut, important leaders get fired. So they have to try really hard to guarantee no failures. Everything takes way longer and costs way more. They still fail sometimes, and just as often a project is cancelled before completion, but that's way better than an explosion, politically.
ghusto | 22 hours ago
AWS et al. (yes, even Microsoft) are far more reliable than the smaller players, let alone your own rig. The flip-side is that when we go down, we all go down together, and I'm fine with that.
blop | 22 hours ago
StanislavPetrov | 19 hours ago
I'll take my own rig 1000 times out of 1000 for reliability. This would be true even if AWS, Microsoft and the rest weren't parasites looking to constantly spy on me.
wallst07 | 13 hours ago
How are you defining it?
jasomill | 2 hours ago
If all you need is shell access and simple tools and have modest security requirements, a few dozen Raspberry Pis (multiple models and revisions) running a stripped down immutable OS image that's only patched for severe bugs with all mutable data replicated between one another and regularly backed up to Usenet should do the trick.
Host them at home, your office, in your car, at the homes of friends and family, etc.
If all your "hosting providers" simultaneously fail catastrophically, you probably have more important things to worry about than your system.
If not, pick up a couple new Pis at retail, pop in SD cards you keep in your wallet containing the most recent base OS image, and grab a latte while your latest backup restores from Usenet over the coffee shop's free Wi-Fi.
aleqs | 18 hours ago
MattSayar | 20 hours ago
JKCalhoun | 19 hours ago
YouTube videos—download.
Music—all stuff I ripped from my own CD collections. Newer digital purchases (Bandcamp) are backed up with everything else.
Magazines from archive.org and worldradiohistory.com—local.
Etc.
yoyohello13 | 19 hours ago
II2II | 19 hours ago
Early Apple history is often spiced up with tales of Woz's exploits using blue boxes, exploiting an inherently insecure centralized communications network.
cbsmith | 18 hours ago
dylan604 | 14 hours ago
cbsmith | 56 minutes ago
Papazsazsa | 14 hours ago
We Millennials mock our parents yet mimic their behavior. Memetic hyper-parasitism.
wallst07 | 13 hours ago
The only thing I saw was memes and news events, 0 direct impact.
anon7000 | 12 hours ago
For most people in the country, despite being online constantly, they would have never known.
If anything it shows that many aspects of the Internet ARE decentralized well, even within one company. I can use a device with apps and accounts that aren’t dependent on the device manufacturer’s uptime.
FrustratedMonky | 23 hours ago
ex-aws-dude | 23 hours ago
thisismyopinion | 23 hours ago
grey-area | 23 hours ago
LoganDark | 23 hours ago
I'm pretty sure it's about the midterms and not about some meeting.
Someone's trying to cause mass panic so they can pin blame.
grey-area | 23 hours ago
Putin does in fact tend to attack interlocutors just before meeting them, he’s done it multiple times to Ukraine.
LoganDark | 23 hours ago
Of course not. My comment was about someone trying to frame them.
> Putin does in fact tend to attack interlocutors just before meeting them, he’s done it multiple times to Ukraine.
If this truly is him, then I wonder how Trump ended up on his bad side all of a sudden.
grey-area | 22 hours ago
I think Putin sees Trump as a tool, nothing more. He has not made his life easy (for example refusing to enter negotiations in good faith on Ukraine, backing Iran etc), if anything Trump seems beholden to Putin somehow (going by his actions like lifting sanctions or supporting russia’s demands in Ukraine), and effusive in his praise of dictators in general (Xi too).
jpatten | 23 hours ago
grey-area | 23 hours ago
If it was AWS fine, that’s not uncommon, but there have been several outages, including at the nsa, which is also unusual.
boutell | 23 hours ago
grey-area | 23 hours ago
alephnerd | 22 hours ago
voakbasda | 22 hours ago
jpatten | 23 hours ago
grey-area | 23 hours ago
teaearlgraycold | 22 hours ago
mixdup | 22 hours ago
Microsoft's Exchange Online outage and the FAA outage had very big impacts, but in the case of the FAA outage wasn't even really that big of a technical issue. Those impacts rightfully were newsworthy but those two plus the PHP fire drill a few weeks ago has everyone on edge
Just because you see a bunch of murders on the news doesn't mean there's a massive increase in violent crime
ShinyLeftPad | 22 hours ago
Was the issue reliably identified then? Last I remember they swore it was definitely not the new SMART thing and failed to provide a satisfactory cause.
mixdup | 22 hours ago
HPE Juniper says the routers were not malfunctioning and did what they were configured to do
The implication is that the router was misconfigured in some way, but no one really knows. Given how government contractors work I buy that story, that they were doing production changes in the network at 3PM on a Friday and had a bad config
No indication at all that it was that new AI tool
lokar | 21 hours ago
I feel like it has to be an issue with whatever routing protocol they are using. So, low traffic in absolute (bps) terms, but something expensive for other routers to process.
I hope they provide a more detailed explanation.
mixdup | 21 hours ago
meeshmuesh | 20 hours ago
lallysingh | 19 hours ago
mixdup | 19 hours ago
You'd be surprised how bad things can be and still skate along without breaking for a long long time
cheesecakegood | 16 hours ago
ShinyLeftPad | 11 hours ago
Except the timing exactly coincides...
mikece | 21 hours ago
lwansbrough | 20 hours ago
FYI you can use, ie. `before:2026-11-01` in Google searches.
wesleywt | 22 hours ago
Reason077 | 21 hours ago
ben_w | 21 hours ago
The code they put up more than 3-6 months ago depending on the provider, I'd agree. But 6 months ago was roughly when the best LLMs got better at spotting issues than humans, and 3 months ago was roughly when open weight models passed the same threshold.
Not that this means everyone's using the best models, of course. Number of times I see academic research and it's "we used GPT-4o and it was bad" and I check the date and oh look one was submitted yesterday: https://arxiv.org/abs/2611.20753
godelski | 22 hours ago
This is clearly HN news. The general news seems to be more concerned about the NSA communication outage
DrewADesign | 21 hours ago
98codes | 20 hours ago
throw0101c | 23 hours ago
In Polish folklore there is apparently a belief that a person must be buried before Sunday (pochówek przed niedzielą?) or 'soon enough', as otherwise it will cause the deceased to pull another family member (or neighbour) into the grave with them. Perhaps the FAA and/or Microsoft outages were not fixed quickly enough?
MiroslavPokorny | 14 hours ago
THey might not have know about germs, but the authorities knew best practices of burying the dead fast and the outcomes of not doing so.
bigfatkitten | 22 hours ago
Outages on classified networks tend to not directly impact the public.
andy_ppp | 22 hours ago
ck2 | 21 hours ago
and screwing up the patches
oh wait, it's EXACTLY that
we just went through PHP having a massive hole that was then found similar in every other language and library
WarmWash | 23 hours ago
Kids were posting videos of their drones on instagram and the nightly news was buying the rights...
sandworm101 | 23 hours ago
>> A US congressman Jeff Van Drew claimed he had heard from "high sources" that the flying objects were tied to Iran. [...] the New Jersey Republican said the drones were coming from an Iranian "mothership" in the Atlantic.
https://www.bbc.com/news/articles/crrwz91wqd9o
reaperducer | 22 hours ago
There are congressmen who believe space lasers change the weather, and a president who thinks COVID can be cured by injecting bleach into his arm. Being elected doesn't make you an expert in anything.
It's used to be amusing to see how much the British press misunderstands American politics and politicians, but more and more it's annoying.
sandworm101 | 22 hours ago
alexandre_m | 21 hours ago
He didn't say that. You lose all credibility by propagating misrepresented statements like this.
Jtsummers | 20 hours ago
https://www.politico.com/news/2021/04/23/trump-bleach-one-ye...
He did suggest, after talking about the effectiveness of disinfects (bleach included in context) on the virus, injecting things into the body. It's very unclear, like a lot of his ramblings, what the hell he was talking about precisely, but it is reasonable to think that he wanted to inject light into the body or disinfectants or both. And if disinfectants, bleach was specifically discussed.
So did he literally say we should inject bleach into arms? No. He didn't mention where the disinfectants or light would be injected, just suggested that they could be.
alexandre_m | 20 hours ago
He was thinking out loud about ideas that could potentially be tested, not instructing people to do anything.
There’s a reason fact checkers rated that claim as false.
It’s the kind of thing people who dislike Trump might joke about over a beer with friends. But bringing it up as a serious argument is ridiculous and intellectually dishonest.
Jtsummers | 20 hours ago
Neither myself nor reaperducer said he instructed anyone, you injected that into the conversation. He did say he thought it could work, literally thought we could inject bleach, other disinfectants, or light into the body to treat COVID. That is not something a sane or competent person would ever suggest, and you even agree he did suggest it.
alexandre_m | 18 hours ago
> thought we could inject bleach, other disinfectants, or light into the body to treat COVID
That's what he suggested, more or less, yes, in the context of exploring treatments. But people are so partisan and obsessed with hating Trump that they'd rather twist his words and make it sound like he was proposing to drink bleach or inject bleach in their arms. Why have nuance when you can have cool talking points.
The full transcript:
"And then I see the disinfectant, where it knocks it out in one minute. And is there a way we can do something like that, by injection inside or almost a cleaning, because you see it gets in the lungs and it does a tremendous number on the lungs, so it’d be interesting to check that, so that you’re going to have to use medical doctors with, but it sounds interesting to me. So, we’ll see, but the whole concept of the light, the way it kills it in one minute. That’s pretty powerful."
brewdad | 20 hours ago
reaperducer | 22 hours ago
Legitimate news organizations don't buy video from the general public. TMZ, yes. New York Times, no.
They only buy video from organizations that have been vetted and with which they have a previous relationship, like a known freelancer. They don't write checks to random kids.
WarmWash | 21 hours ago
interestica | 21 hours ago
dgellow | 23 hours ago
MiroslavPokorny | 14 hours ago
EGreg | 23 hours ago
NSA
Chase and banks
Airlines
Yeah, only one thing connects them: all being unprecedented and in the same week. Nothing to see here folks!
testrun | 22 hours ago
za_creature | 22 hours ago
Who is to say what said company might have done with anything it learned in the process?
fourside | 22 hours ago
nvme0n1p1 | 21 hours ago
Maybe this hundred-year storm happened because the gods are angry with you. Or maybe it happened because a hundred years* have passed. You'll need some evidence for that first one.
* on average
za_creature | 21 hours ago
is manufacturing consent.
EGreg | 19 hours ago
If these are all coincidential then we should be able to multiply the probabilities. All in all we have at least 10 different companies — including Google — with systems down. When was the last time this occurred in the same week?
https://www.dailymail.com/news/article-16299951/amp/american...
1. Microsoft (Outlook, Teams, Copilot, Microsoft 365) — Nov. 12–14, 2026: A global cloud and network capacity issue disrupted email, messaging, and workplace apps for millions of users before restoration on Saturday.
2. Federal Aviation Administration (Air Traffic Control Communications & Radar) — Nov. 13, 2026: A telecommunications and router data-flooding issue triggered a multi-hour nationwide ground stop and thousands of flight delays and cancellations.
3. JPMorgan Chase (Online Banking, Transactions & ATMs) — Nov. 16, 2026: A software bug in an on-premise hardware system slowed transactions and caused login and payment issues for customers.
4. Zelle (Peer-to-Peer Payment Transfers) — Nov. 16, 2026: Users reported payment delays and errors tied to bank processing issues.
5. Wells Fargo (Online Banking & Account Access) — Nov. 16, 2026: Customers reported account access and transaction issues during the broader banking disruption.
6. Bank of America (Online Banking & Account Access) — Nov. 16, 2026: Customers flagged account access and transfer problems amid the Monday banking disruptions.
7. U.S. Bank (Online Banking & Zelle Transfers) — Nov. 16, 2026: Users reported access issues and potential delays on cross-institution Zelle payments.
8. United Airlines (Flight Departure Systems) — Nov. 16, 2026: A technical failure affecting departing flight operations caused delays for more than 90 minutes.
9. Apple (iCloud Sign-In, App Store, Apple Music, AppleCare & Apple Pay) — Nov. 17, 2026: An internal update caused outages and sign-in issues across multiple Apple services.
10. Marriott International (Marriott.com & Marriott Bonvoy App) — Nov. 17, 2026: A technical issue disrupted sign-in capabilities on the website and mobile app before resolution.
11. Zoom (Website / Optimizely Vulnerability) — Nov. 17, 2026: Affected by a security incident involving malicious code injected via the Optimizely platform across major company websites.
12. CNN (Website / Optimizely Vulnerability) — Nov. 17, 2026: Among major sites impacted by the Optimizely malicious JavaScript incident.
13. BBC (Website / Optimizely Vulnerability) — Nov. 17, 2026: Reported among major international sites affected by the Optimizely exploit.
14. Salesforce (Website / Optimizely Vulnerability) — Nov. 17, 2026: Listed among major websites impacted by the Optimizely security incident.
15. IKEA (Website / Optimizely Vulnerability) — Nov. 17, 2026: Included among major company websites affected by the Optimizely incident.
nvme0n1p1 | 18 hours ago
NeutralCrane | 20 hours ago
yunwal | 18 hours ago
Point to a week last year that was on par with this one and I'll believe you
ajasmin | 20 hours ago
zarzavat | 22 hours ago
ryeights | 22 hours ago
zarzavat | 21 hours ago
ryeights | 20 hours ago
za_creature | 20 hours ago
Have we now? I guess it must be those rotten humans that were fired, being all petty and hacking the perfect systems written by this new competence.
za_creature | 21 hours ago
Relying on code generation has pretty much been proven to atrophy skills in the short term and relying on opaque internet-connected black boxes for code generation is both leaking information and inviting back doors, should that black box be adversarial (whether due to the survival incentives of an intelligence or more likely due to the intentional biasing by the ones in control of the box).
freediddy | 22 hours ago
JohnMakin | 22 hours ago
gorgoiler | 22 hours ago
The original poster is claiming this is a news cycle perversely creating the (perception of a) phenomenon itself, instead of the more logically-straightforward but incorrect assumption that this is a phenomenon being picked up as a news cycle.
The winner of this battle of Internet commentary will of course be the one who brings statistics and outliers to the fight. Gentlemen/women/folk: stand back to back, take ten paces and then let fly your volley of anecdotes, data, or actual reasoning, if you please.
JohnMakin | 21 hours ago
In fact, I find the sudden rush to explain away such events as more suspicious than the actual events themselves.
Eji1700 | 22 hours ago
That said outlook being down for 24 hours, all flights being grounded, and apple not being able to do ANY payments would all make headlines on any given day. This isn't just "now that the populace is focused on it we'll report anything remotely related".
That said it does just seem to be a mix of bad practices catching up with bad decisions being made (hey lets update on friday..), but even I have to agree the frequency and duration is starting to look odd.
That said as with all these things, I sincerely doubt the ability of anyone to keep this shit under wraps. Only the US and its companies are being targeted AND no one from those companies screamed bloody murder about coordinated attacks/rogue AI before the gov stepped in and said "shut the fuck up."?
Maybe, but I doubt it? I know people are at risk of losing their jobs, but word gets around and I think there'd be a hell of a lot more discussion about possibilities by people who actually know what the fuck they're talking about, rather than talking heads and random nobodies who clearly don't know the industry (as often happens when any industry gets signal boosted by the media).
Hanlon's razor/Gellman amnesia still seem like the strongest answers mixed with maybe some "yeah so MS shitting the bed helped kick off a chain of tech dominoes"
jlarocco | 12 hours ago
Eji1700 | 12 hours ago
hammock | 21 hours ago
Frost1x | 21 hours ago
It’s undoubtedly a good perspective to be aware of and understand.
At the same time, we should also assume if we don’t have information to prove things aren’t correlated or reasonable evidence to think they are unrelated, we shouldn’t just dismiss them. Sometimes, correlation exists because of an underlying casual relation. It’s not always, there’s not even some frequency I could reasonably assign. So we shouldn’t dismiss it either.
I keep seeing the two polarized perspectives with few considering that it may be a mixture. Some of these events really could be related. Some may actually be pure coincidence. Some may be simply Baader–Meinhof (frequency illusion) related (we see something, therefore we look for it more, therefore we see more of it, so on) as a survival instinct.
I don’t know if there’s clarity what the current mixture is. I can read details about several of the incidents reported causes and say clearly they’re unrelated but I’m not entirely sure I trust some of the reports as it’s not in their interest to say something malicious did cause these problems.
whalesalad | 23 hours ago
Personally I think there's a more likely explanation.
The arrival of sophisticated frontier models like Mythos is giving us an explosion in vuln discovery. CVEs, zero-days, critical infrastructure bugs, etc. There's a massive backlog of systems that suddenly need urgent attention/patching and a lot of this crap has probably been sitting out of sight and out of mind for years or more.
A tiny little security patch can cascade into major OS upgrades, TLS incompatibilities, dead protocols, dep conflicts, cert issues, and all kinds of unexpected noise. Multiply that by thousands of org (many of which are understaffed and running ancient infrastructure) and you got a recipe for chaos.
I think about the original meaning of "catfish" - long haul ocean carriers would put catfish in tanks with other fish to keep them agitated and moving so they wouldn't deteriorate during transport. AI security research is a catfish that is forcing an enormous amount of long-overdue infrastructure maintenance.
Could it be a coordinated attack? Sure. But I am leaning towards just general lack of IT competency that is getting some sunlight due to Mythos/Glasswing/etc
basch | 23 hours ago
It’s the chaotic good of conspiracy theories we can use to imagine a world where there’s a plan and things are going according to it vs rudderless happenstance.
CoolestBeans | 23 hours ago
tedd4u | 22 hours ago
CoolestBeans | 22 hours ago
ryeights | 22 hours ago
apercu | 23 hours ago
whalesalad | 22 hours ago
nalekberov | 22 hours ago
It reminds me of the guy who was banned by Apple for the wrong reason [1]. It took him several days, and he had to reach important people at Apple, to get his account back.
[1] https://hey.paris/posts/appleid/
exographicskip | 22 hours ago
Apple certainly makes that hard to do, especially if you publish any apps on their devices. Or use any of their services that have tight integration in their ecosystem.
user2722 | 22 hours ago
Giant LLMs are being deployed as a police/judge of traffic on the edge of important companies. You can't deploy slowly because China!
mkhalil | 22 hours ago
https://developer.apple.com/system-status/
https://www.apple.com/support/systemstatus/
Interesting that HomeKit makes the general status page and not the developer one. Makes me wonder how they define each entry, and what makes up a system in one vs. the other.
trickypr | 22 hours ago
A normal user might expect to access their home devices from outside the house (which is what I presume "HomeKit" covers). But only a "publisher" (for want of a better term) would need to access App Store Connect, or News Publisher or similar.
joshstrange | 22 hours ago
And no one noticed because it was just as broken as always?
Honestly, Apple Developer and App Store Connect are miserable to work with as is the whole Apple review process.
DenisM | 22 hours ago
AI race redirected resources and attention away from operational concerns. There’s also seasonal increase in traffic in many industries, which may put extra pressure to deploy updates faster.
dgggh | 22 hours ago
andrewla | 22 hours ago
buildwrangler | 22 hours ago
Sarkie | 22 hours ago
rzerowan | 22 hours ago
kombine | 22 hours ago
StanislavPetrov | 19 hours ago
dongliliu | 21 hours ago
alexandre_m | 21 hours ago
Time to withdraw your money /s
amir_ghm | 21 hours ago
m3kw9 | 20 hours ago
dboreham | 20 hours ago
rootusrootus | 19 hours ago
dannyw | 8 hours ago
They were having IT issues, so they just wrote something down on a piece of paper, went to grab something from the warehouse, noted the SKU/serial number, and gave it to me.
mrcyna | 20 hours ago
dmix | 20 hours ago
The app store is under a lot of pressure from tons of AI-written mobile apps being submitted. Similar to Github.
blobbers | 20 hours ago
boothby | 18 hours ago
NoPicklez | 17 hours ago
These interruptions are potentially a combination of companies racing to patch faster to keep up, or a large exploit has been found, or importantly AI exploit tools like ARTEX are being used in the wild.
I have nothing else to go on
octoberfranklin | 17 hours ago
cush | 12 hours ago
freakynit | 14 hours ago
Animats | 14 hours ago
dylan604 | 14 hours ago
KingMob | 12 hours ago
Frankly unsure if Apple Pay is worth the bother for us, at this point.
treebeard901 | 10 hours ago
The number of vulernable systems far outweighs all of the patching being done by the defenders. Exponentially increasing hacks and unexplained outages begin occurring in many systems.
AI patching turns out to not only cause the initial downtime from the accrued technical debt, in many situations the rush to put out a patch for a known exploit itself causes more outages because AI is not advanced enough to have enough context for every system it is trying to fix.
But so many of these systems can't be taken offline because the modern world depends on them. The only solution is going to be to destroy the open internet, strict firewalls between countries and regions, and excessive internal surveillance to track down any domestic threats.
The open/global internet will have to get locked down.
tomaskafka | 5 hours ago
If it’s actually red, that means things are so bad Apple can’t mask it anymore.
(btw, there was a post going around guessing that LLMs has discovered 0day with exploits in the wild so serious, that several huge companies took their stuff offline to fix it)