Anthropic was using fearmongering to sell by calling Mythos dangerous per @GaryMarcus
But in Bloomberg interview Dario says it had hurt them financially to not release it publicly
@KobiHackenburg@AISecurityInst@UniofOxford This is very very concerning. Frontier models will only get better and better. We know how social media companies will use them,
Anthropic is yet to fully unlock self-improving AI.
If it does, there's gonna be intelligence EXPLOSION๐ฅ
Timestamps:
0:00 Intro
0:46 Recursive Self Improving AI
1:58 Claim 1
3:43 Claim 2
3:53 Claim 3
4:11 Claim 4
4:44 Claim 5
5:41 Claim 6
6:02 Claim 7
7:17 The problem with AI alignment
7:23 Automating AI development itself
8:07 Already predicted in AI-2027
8:31 Anthropic is highly uncertain about the impact
9:13 Anthropic calls for AI SLOWDOWN!
11:15 Catastrophic risks
12:10 Questions remain
12:40 Slowdown VS Race in AI-2027
Researchers show that Claude Code is 98% not AI.
Anthropic never gave us the architecture for Claude Code. There were no docs. Just a tool that every developer is currently obsessing over.
Until it leaked recently.
A research team pulled the source code, analyzed all 500,000 lines, and found something ridiculous.
Only 1.6% of the codebase actually interacts with the AI model.
The core of Claude Code is literally just a simple while-loop. It asks the model what to do, runs a tool, and repeats.
So what is the other 98.4%?
It is hardcore, traditional software engineering.
The researchers found a massive, complex infrastructure designed entirely to babysit the AI and keep it from hallucinating or destroying your computer:
- A 7-mode permission system acting as a security bouncer.
- A 5-layer context compaction pipeline so the AI doesn't forget its goal.
- A subagent delegation mechanism with strict worktree isolation.
- Four different extensibility hooks to manage external tools safely.
Every startup right now is trying to build a better AI model to get better results.
Anthropic did the exact opposite.
They took an existing model and built a fortress of deterministic software around it.
They realized that the AI doesn't need to be smarter. It needs to be managed.
Anthropic might have just unlocked the Secret to Intelligence Explosion ๐คฏ
Tech meditator breaks down:
1) What is Recursive AI development
2) Have they really unlocked the Secret to intelligence explosion?
3) What are the basis of their claims?
4) What are the implications of this recursive AI development?
โIf an AI cannot apply an abstract lesson to a new situation, it is not truly reasoning or learningโ, new study with further evidence backing up what I have been saying for 25 years.
cc @dwarkesh_sp
Fable isn't the first.
In 1999 the department of defense blocked exports of the PowerMac G4 for crossing the 1 gigaflop threshold.
Steve Jobs turned it into an ad.