AI Socratic Sep 2025 Part 3 — Frontier Tower Edition

This blog post has the goal to help us structure the Socratic dialogues at the AI Dinner 13.0, that Ivo and I are organizing in SF at the Frontier Tower.

We'll use blog post 1 and part 2 to go through Since part 2 was written during the last AI dinner 13.2 and a lot of new updates happened since, we'll cover them here.

#Why Language Model Hallucinate and Cannot Say “I don’t Know”!

Language models hallucinate because their training and evaluation reward guessing over admitting uncertainty. Models are unable to say “I don’t Know” because they focus on accuracy. Guessing can improve scores but leads to more confident errors (hallucinations).

Some facts are unpredictable or unavailable, making errors inevitable. The solution might be roughly penalizing errors more than uncertainty and rewarding honest abstention.

https://openai.com/index/why-language-models-hallucinate

#How people are using ChatGPT & Claude

OpenAI just released how people are using ChatpGPT. Coding 4.2% We live in a bubble

https://openai.com/index/how-people-are-using-chatgpt, https://x.com/atelicinvest/status/1967941812078915614

Claude also released how people are using their API and Claude.ai. The 1P API is used at 97% for automation.

#Google AP2 + Coinbase x402, A2A stablecoin payments

AP2 (Agentic Payment Protocol) and x402 just unlocked a new level for AI agents. Agent can now actually pay other agents and MCPs.
https://x.com/sundarpichai/status/1968013016181641492

#Oracle Up 40% Because It’s building Stargate

https://x.com/mattturck/status/1966545601140531707

#Full Sources List

AI For Builders

• Writing Effective Tools for Agents https://www.anthropic.com/engineering/writing-tools-for-agents
• MCP support in chatgpt https://x.com/gdb/status/1965810388966248652
• Most important chart on codex https://x.com/swyx/status/1967651870018838765
• GPT-5 dominates the long horizon agent race https://x.com/slow_developer/status/1967186229775995305

DeAI

• x402 + MCP + AI SDK https://x.com/ethanniser/status/1966550369657299123
• ETH Foundation introduces dAI team https://x.com/ethereumfndn/status/1967579790938099988

Founding

• Not a bubble https://x.com/Speculator_io/status/1966623422978424858
• Oracle 40% up https://x.com/mattturck/status/1966545601140531707

Lol

• Shaved my head wdyt https://x.com/alxfazio/status/1966497162641940629
• Writing effective tools for llm agents https://x.com/AnthropicAI/status/1966236220868247701

Opinion

• The ultimate instrument of power is the control of the objective function https://x.com/EMostaque/status/1967917059783938555
• How cursor and windsurf survive when every hyperscalers is releasing their own CLI tool https://x.com/BenjaminDEKR/status/1967815927120212281
• future of AGI is closed source https://x.com/MechanizeWork/status/1967681892586860784
• Apple is not behind https://x.com/mattcassinelli/status/1967233259483648170
• If you’re worried about AI, I’ll say it again: AI won’t take your job. It will let you do ANY job https://x.com/PeterDiamandis/status/1967348046716346521

Random

• ⭐ Chinese models download on hugging face https://x.com/OmerCheeema/status/1967308542836445572
• Apply for jobs automatically https://x.com/aidancramer/status/1967888482917027909
• Country of geniuses in a data center https://x.com/sundeep/status/1967755262242197756
• scraping tiktok to frontrun political prediction markets https://x.com/MovieTimeDev/status/1967725247072964726

From gears to gradients. the single most influential trajectory of all time.

• Pick 3 (Ivo in this picture) https://x.com/awnihannun/status/1967299162380357701

• Karpathy https://x.com/karpathy/status/1966896849929073106
• List of 2000 orgs https://x.com/Scobleizer/status/1966984813765947856
• meta uses AI for recruiting https://x.com/vxunderground/status/1966640516436574467
• OpenAI consider moving to NYC https://x.com/JakeZegil/status/1966197843217305609
• There is no word in English that means someone is 30% sure something will happen https://x.com/AJThurston/status/1965724458251059695

Research

• “My Boyfriend is AI” MIT study on reddit 27k community https://x.com/arankomatsuzaki/status/1967812112887255055
• Google research: An AI system to help scientists write expert-level empirical software https://x.com/shaneguML/status/1966869683749220748
• LLMs just learned how to explain their own thoughts. https://x.com/VraserX/status/1967152142323433551
• Language Models Are Capable of Metacognitive Monitoring and Control of Their Internal Activations https://x.com/Sauers_/status/1967045339783033203

Visuals

• https://x.com/wilplatypus/status/1966897410593919278
• https://x.com/NTFabiano/status/1966099820139131297

Video

• World Lab world model from a picture https://x.com/venturetwins/status/1968109391217127701, https://x.com/CoinbaseDev/status/1967966011833061749