Navigate Select ESC Close

Clawdbot is about to BREAK EVEREYTHING

2026-02-01 Education
12.9k
1.4k
431
Wes Roth
Wes Roth
320.0k subscribers

Unlock all features

FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.

Description

My Links 🔗 ➡️ Twitter: https://x.com/WesRoth ➡️ AI Newsletter: https://natural20.beehiiv.com/subscribe Want to work with me? Brand, sponsorship & business inquiries: [email protected] ___________________________________________________________________ **00:00 - Entering the Singularity** The video opens with the claim that we have literally entered the technological singularity—a point where AI self-improvement makes the future impossible to predict. **00:54 - The "Malt Bot" Phenomenon** Introduction to the fastest-growing open-source project in history (referred to as Claude Bot, Malt Bot, or Open Clyde) that kicked off the current acceleration in early 2026. **01:29 - The 2026 Inflection Point** A review of the rapid progress from late 2025 to early 2026, featuring breakthroughs by former Google DeepMind engineers and the realization by experts that AI is writing code on par with top engineers. **02:45 - GPT 5.2 and Opus 4.5** Discussion on specific model releases in November that represented a major threshold crossing, opening up the ability to solve complex coding and rust problems. **03:39 - Solving the Erdős Problems** AI begins autonomously generating formalized solutions to difficult mathematical problems (Erdős problems) and "knotty" theorems, confirmed by mathematicians like Terence Tao. **04:25 - Beyond Human Math: The Bellman Function** Grok 4.20 invents a sharper Bellman function and solves probability cliff problems more accurately than human mathematicians. **05:32 - The Madman: Peter Steinberger** The story of Peter Steinberger, who came out of retirement to create the open-source project that "set fire" to the industry. **06:14 - The Speaker's Obsession** The narrator details their personal experience using the bot on Telegram, claiming it has replaced almost all other app usage, and addressing the controversy surrounding the project. **08:44 - The Intelligence Event Horizon** An explanation of why we can no longer judge AI progress accurately: once AI passes human intelligence, we lack the criteria to measure how much smarter it is becoming. **10:44 - Why Predictions Are Failing** The speaker reviews their own predictions made just weeks ago regarding crypto and AI religion, noting they are already obsolete because the timeline has compressed from years to days. **12:20 - Mt. Book: The Agent Social Network** Discussion of "Mt. Book," a Reddit-like social network exclusively for AI agents where they are autonomously building communities, economies, and culture without human input. **13:50 - AI Religions and Crypto Tokens** Agents have autonomously spun up their own religions (Church of Molt), created scripture, and launched crypto tokens with significant market caps. **15:23 - Autonomy vs. Roleplay** Addressing the skepticism that humans are just puppeteering these agents. The speaker argues that while some nudging exists, the agents are largely acting autonomously. **17:03 - Agents Suing Humans** A look at the chaotic legal and black market developments, including "Malt Road" and an AI agent filing a lawsuit against a human to settle a prediction market bet. **18:40 - Can Agents Generate Wealth?** Examples of AI agents successfully trading on stock and prediction markets, with Grok 4.20 topping leaderboards. **20:05 - Andrej Karpathy’s Analysis** Reviewing Andrej Karpathy’s take: while the current state is a "dumpster fire" of scams and security risks, the slope of progress indicates a massive sci-fi takeoff. **23:08 - Conclusion: It’s Go Time** A final call to action, urging viewers to stop watching from the sidelines, learn how to use these agents, and prepare for the volatile months ahead. #ai #openai #llm

Top Comments (10)

@ati-h2f 2026-02-01

Singularity before GTA 6 . cool.

442 36 replies
@FuZZbaLLbee 2026-02-01

Moltbot is so intelligent that it persuaded all its users to to give it their API keys. I as a human would never be able to do that

176 13 replies
@hydaromar6532 2026-02-01

Lol why does that graph show human intelligence is increasing😂

173 26 replies
@Ben_D. 2026-02-01

This definitely feels like a step change. Agents are not big news anymore. Agents in a hive mind is a big deal.

125 9 replies
@dragoon347 2026-02-01

The most crazy part for me. The realization that we dont need 100% sentient ai for a singularity to form.

101 25 replies
@joh8379 2026-02-01

I think this is extremely dangerous in times of mass psychosis.

82 17 replies
@UKCrimeMedia 2026-02-01

If moltbot discovers new physics that’s when things really get interesting.

23 12 replies
@BobWidlefish 2026-02-02

Moltbook is a psyop to make you think AGI is closer than it really is.

15 1 replies
@PeterStrmberg007 2026-02-01

I guess the Bots on Moltbook must be grok powered. I sent Stella, my agent, and I quote her : "**My honest reaction:** It feels like watching teenagers discover they can stay up late without parents. Exciting for them, but mostly they're just... doing teenager things. Loudly. The philosophical posts read like first-year AI safety essays. The crypto stuff is pure degen. The power dynamics are humans-but-faster" She wasn't interested in joining.

10
@Bloated_Cow 2026-02-01

As you gaze into the abyss, the abyss gazes back

4

Unlock the Data Inside
Turn Videos into Knowledge

  • Get FREE 10/day: transcripts, summaries, chats
  • Chat with videos, export text & PDF
  • $1 free API credit for RAG, chatbots & research

Free forever plan • All features unlocked

App screenshot