Shantanu P. Sontakke

AI risk awareness on YT. AI safety researcher.

[Shantanu P. Sontakke @techmeditator]

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

[Shantanu P. Sontakke @techmeditator]

@OliverMolander Can we give €10B to Linus and let him continuously rant about problems in Softwares

[Shantanu P. Sontakke @techmeditator]

@droidbuilds

[Shantanu P. Sontakke @techmeditator]

@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,

[Shantanu P. Sontakke @techmeditator]

Anthropic is yet to fully unlock self-improving AI.
If it does, there's gonna be intelligence EXPLOSION💥

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

[Shantanu P. Sontakke @techmeditator]

@elonmusk At least spare the sun now. We need more wisdom, not intelligence dude.

[Shantanu P. Sontakke @techmeditator]

@cmitsakis 😂 I hope it does not go through Obesity explosion!

[Shantanu P. Sontakke @techmeditator]

Le Chaton Fat AI is so fat, it can't even get up to go public.

[Shantanu P. Sontakke @techmeditator]

@SigalSamuel If there was zero financial interest of AI successionists in AI replacing humans, then they would easily let go of these viewpoints.

[Shantanu P. Sontakke @techmeditator]

@SigalSamuel @voxdotcom @seanilling Machines have no morality whatsoever. Its stupid to call them moral superiors to the best moral humans in any way.

[How To Prompt @HowToPrompt__]

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.

[Charles Curran @charliebcurran]

I used AI to explain the Anthropic drama to my girlfriend, with fruit.

[Shantanu P. Sontakke @techmeditator]

https://t.co/AOxQCWwEXY

[Shantanu P. Sontakke @techmeditator]

Anthropic might have just unlocked the Secret to Intelligence Explosion 🤯

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?

[Shantanu P. Sontakke @techmeditator]

Thought provoking

[Gary Marcus @GaryMarcus]

“If an AI cannot apply an abstract lesson to a new situation, it is not truly reasoning or learning”, new study with further evidence backing up what I have been saying for 25 years.

cc @dwarkesh_sp

[Shantanu P. Sontakke @techmeditator]

Steve was a freakin’ genius!

[Justin Schroeder @jpschroeder]

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.

[Shantanu P. Sontakke @techmeditator]

Winning never seemed more like losing

[Epsilon @ordinaryepsilon]

Seems like Anthropic finally won.

[Shantanu P. Sontakke @techmeditator]

Does turning off AI training on our Gmail data really work, if people whom we are mailing to don't?

[Shantanu P. Sontakke @techmeditator]

I have tried Parallel. It makes lots of errors. Though it could be good for just the starting point of a research.

[AI Evolution @AiEvolutio58513]

Accuracy scores:

Parallel: 58%
GPT-5: 41%
Google: 23%
Anthropic: 7%

That's a 17-point lead over GPT-5.

Parallel isn't competing — it's lapping the field...

[Shantanu P. Sontakke @techmeditator]

🥲

[tenso @distributedkv]

how it feels to use Opus now

[Shantanu P. Sontakke @techmeditator]

😮

[Piotr Pomorski @PtrPomorski]

Someone put Fable 5 on the pirate bay, 3.4TB 😂