Remix.run Logo
▲ MrBuddyCasino 17 hours ago

What happen right now is that every organization on earth is under attack from automated LLM bots. This is why we are currently seeing so many outages. Nobody admits that this is the case.

▲alchemist1e9 17 hours ago | parent | next [-]

Everyone is assuming it’s LLM bots but it apparently isn’t.

https://www.washingtonsun.com/technology/bank-ceos-nsa-tell-...

> Bank CEOs, NSA Tell Congress That AI-Driven Patches Are Behind Outages

> Artificial intelligence is not causing the disruptions but identifying gaps that are difficult to fix.

▲theptip 17 hours ago | parent [-]

It’s an interesting angle, but this is basically the same cause as GP.

The base threat level has increased dramatically. Now everyone is racing to use Glasswing/Daybreak/whatever to fix the bugs before someone external finds them.

So yeah, maybe not a direct result of bots attacking, but still an indirect one.

▲autoexec 15 hours ago | parent [-]

I think there's a big difference between malicious actors using AI bots to break everything and multi-billion dollar corporations and governments vibe coding patches that break everything. You can't do much to stop AI bots from testing your defenses, but pushing a bunch of untested updates to everything in production in the middle of a busy day is a choice.

These kinds of outages will just point out which idiots have no business managing these systems in the first place and they'll hopefully all be fired so they can be replaced by people who know better.

▲theptip an hour ago | parent | next [-]

Sort of. The key observation is that every company now needs to move faster. A bunch of companies were just good-enough, and now on the margin they are not good enough.

I don’t think it’s as binary as “bad companies pushing dumb fixes”. Everyone is getting squeezed, and everyone is making more mistakes. Some worse than others though!

This is an opportunity for the better companies, but certainly disruptive for everyone in the meantime.

▲balex 14 hours ago | parent | prev [-]

This isn't a solved issue. People don't know better yet. That's why there's so many startups in defensive AI.

▲anon373839 16 hours ago | parent | prev | next [-]

I am very reluctant to buy into that narrative because it's just a bit too convenient for a certain AI lab that desperately needs a strong IPO and has no path to justify its valuation if it can't find a way to monopolize the technology.

▲MrBuddyCasino 14 hours ago | parent [-]

Do I have to add a “fuck Anthropic” footer to every comment.

▲brookst 13 hours ago | parent | prev [-]

Who do you expect to “admit” this? It’s common knowledge. Everyone on HN knows it. It’s like “admitting” that the sun rises in the east.