Fact◆Article · 36 words
Meta's AI bot allowed anyone to take over any account simply by asking it to change that account's email address, with zero authentication required.
Gergely Orosz · The Pragmatic Engineer
ClaimVideo · 16:15 — 17:45
AI labs are relying on iterative deployment and defense-in-depth safeguards that have worked so far, but nobody knows if or when these paradigms will break, and they could fail at especially critical moments.
Nathan warns that the iterative-deployment and defense-in-depth approach that has worked for AI labs so far is untested at higher capability levels, and could fail precisely when the stakes are highest. ✦ AI generated
Nathan · The Cognitive Revolution · 2026-07-02 · original ↗
starts at this moment · 16:15
But you do wonder I mean all these models kind of start to come under some strain as you get to sufficiently powerful capabilities or or regimes where a small enough gap in the defenses is enough to create a huge problem. And yeah, it feels like in in multiple ways. We're kind of riding these paradigms that have worked well so far and we just don't know if and when they might break and if they do break it might they might be breaking at kind of critical times which is a strange juaposition on on multiple different levels.
verbatim transcript · starts at 16:15
Transcript · around this moment
16:15deployment is kind of the best path to uh um things going well broadly. So I think that still makes sense. But you do wonder I mean all these models kind of start to come under some strain as you get to sufficiently powerful capabilities or or regimes where a small enough gap in the defenses is enough to create a huge problem. And yeah, it feels like in in multiple ways.
16:50We're kind of riding these paradigms that have worked well so far and we just don't know if and when they might break and if they do break it might they might be breaking at kind of critical times which is a strange juaposition on on multiple different levels. People talk about that all the time of course with alignment but I think it extends to these defense in-depth safeguards and it it
17:16def I would you know at the highest level it kind of extends to the whole uh iterative deployment model. So as always confusing um but selfishly I'm glad to have it back that's for sure and I am excited to check out uh 5.6 as well. I feel like it's probably going to be for multiple reasons. One, obviously, they're going to start charging. We'll see. I mean, I kind of expect it's going
17:44to be interesting to watch Anthropic dance around limits and usage because OpenAI is going to offer a ton more tokens, it seems, with 5.6 at their $200 price point than Anthropic is going to be offering for Fable with their $200 price point. As of now, from what we have seen with the the relaunch, it goes until July 7 on this preview access basis and then you have to pay
18:14the API rates and that's going to be order of magnitude, you know, maybe maybe more than an order of magnitude more expensive than 5.6 tokens. >> Yeah. >> If you have the pro subscription. So, it's going to be interesting to watch how they manage that. I would I would bet that Fable comes back to the Claude Max subscription. It seems just very hard for them to have no Fable
Related moments
Claim◆Article · 52 words
Banning open models now while closed models race 10-100X ahead in the hands of one or two companies creates a bigger long-term risk than allowing open Mythos-class models to exist.
Nathan Lambert · Interconnects
Prediction◆Article · 43 words
AI systems are expanding their economically relevant capabilities faster than humans are expanding their comparative advantage relative to AI, meaning the economy is likely headed toward person-light or person-nil, AI-heavy organizations outcompeting unaugmented humans rather than staying fundamentally the same.
Jack Clark (Import AI) · Import AI