It was just another Tuesday, until it wasn’t. 20% of the internet vanished in a heartbeat, all because Cloudflare, “which is supposed to protect the internet from attacks, accidentally “attacked” itself… when a  routine configuration change (database permission update) triggered a hidden bug in its bot protection system, and in an instant, this “gatekeeper” locked everyone out,” Bitget reported.

It’s not the first major outage we’ve witnessed in the past few months. “Amazon’s AWS Goes Down, Takes Out “Half of the Internet,” said Futurism. “Apps and platforms relying on Amazon Web Services (AWS), Amazon’s cloud computing service, were in a jam… in a striking example of how infrastructure consolidation makes the modern internet vulnerable to a failure by a single major provider.”

And this keeps happening. Days after AWS went dark Microsoft Azure experienced a major outage. “Cisco’s network monitoring service has logged 12 major outages in 2025 so far, 23 in 2024 and 13 in 2023. Cloud service provider outages climbed from 17% to 27% of all outages in 2024, while ISP outages decreased from 83% to 73%,” the Times of India reported.

And so much for too big to fail. The failure rate is getting worse, and the problem is the now centralized web. Well, that’s one of them, at least.

At the start, the internet was about decentralization. It was designed to survive a nuclear war: if one part became compromised, the system would merely reroute itself. Along came consolidation in the form of AWS, Google Cloud and Azure, and why not? Bigger, faster, better, cheaper, right? The problem is, these services created a problem the internet was designed to avoid: centralization. As we’ve witnessed, when these services go offline, even if only temporarily, they take the internet down  with it, which is a major problem considering that even essential services, like 9-1-1, have gone electronic in the name of cost-savings. Closer to home: “AWS outage reminds us why $2,449 Internet-dependent beds are a bad idea, said Ars Technica. “The reliance on the Internet for smart bed products from Eight Sleep resulted in people being awoken by beds locked into inclined positions and sweltering temperatures,” aka, the Internet of Things that Could Go Wrong. Quickly.

Note to self: the blockchain was functioning just fine during the Cloudflare outage, except, users are dependent on a Web 2 layer to access it, so nice idea but bye-bye mon cowboy.

“Today’s internet is also built on an aging infrastructure. The “AWS and Azure outages recently were both due to DNS which is a bigger problem, IMO. DNS is very old tech, and the people who know it deeply and can wrap their brains around a very tangled global web are aging out of the workforce. Those of us with deep tribal knowledge are being told we’re being replaced by AI. As I approach retirement age, I’m going to sit back… and watch as the youngins with zero exposure or experience with DNS fumble with it when a serious outage occurs taking us down for days, not just hours,” observed one commenter re the Cloudflare outage.

“After Cloudflare went down, I asked an AI program to dig into how connected everything really is. Turns out we rely on computers, the internet, and AI for more parts of everyday life than I realized. Nothing in human history has ever been this linked together, and it made me think about what would happen if that link snapped.”

Speaking of which, we’ll soon be facing the Y38 bug. As Wikipedia notes, “The year 2038 problem (also known as Y2038, Y2K38 superbug, or the Epochalypse) is a time computing problem that leaves some computer systems unable to represent times after 03:14:07 UTC on 19 January 2038.”

“The hype surrounding Y2K was because no one knew what could really happen. It ended up being a relatively simple — though time-consuming and costly — fix because programmers knew exactly what the problem was and how to solve it. In the future, if a major system goes down, they might not know what the problem is or how to fix it. That’s the scary part,” the previously-cited commenter continued.

Finally, there’s the part of all of these recent – and growing –  major outages that not many are talking about. Namely, the enormous power over our lives held in the hands of very few. As Futurism noted re the AWS outage, “Really shows how easy it would be for Bezos and Ellison to just turn off the internet if they wanted to, for any reason. “In fact, the”global cloud infrastructure — the systems hosting the websites and services that fuel our economy — is overwhelmingly controlled by two companies, AWS and Microsoft’s Azure. (Google holds a distant third.) Although, “last quarter, Alphabet’s cloud revenue jumped 32%,” Morning Brew noted.

Your focus, founders et al, should be what are your contingency plans/redundancies when – not if – the internet next goes down, and can you access them without a screen? That it will go down again is money in the bank. The real question is, when it occurs, and speaking of contingencies, will you be able to get at least a bit of your banked money out of it? Onward and forward.