AI, three at a time · read · watch · listen

Tag

recursive-self-improvement

2 reads

In Good Company with Nicolai Tangen 2026-10-21-2

Eric Schmidt: The Age of Agents, Superintelligence, and Why Europe Is Falling Behind

Schmidt says the gains are real but concentrated in software, cyber and math, where learning loops are fast, and biology lags on data. The effect he is most confident about is organizational: a friend's firm cut 20% of staff, the 'glue' planners, and an estimated 2,000 of 10,000 people at a large firm are unneeded. Read against deBoer's cancer example, displacement arrives first where the work is coordination, not where the problem is physical.

# tags
in-good-companypodcastagentic-aiorg-designchange-managementeuropean-aieuropeai-regulationchina-aidistillationopen-modelsai-geopoliticsrecursive-self-improvementsuperintelligenceworkforce-dynamicsai-bubblesovereign-wealth-funds
◆ entities
Recursive self-improvementTeacher-student distillationOrganizational absorption capacitySuperintelligenceAI regulatory risk
→ threads
ai-regulationai-bubble
⟷ links
2026-05-27-22026-04-03-22026-03-13-w32026-04-10-12026-03-08-12026-04-12-32026-04-05-12026-05-01-32026-03-30-32026-04-17-w3
permalink
OpenAI 2026-05-20-3

OpenAI Model Disproves Erdos Unit Distance Conjecture

An internal OpenAI model disproved Erdos's 1946 planar unit distance conjecture, with Princeton's Sawin extracting an explicit exponent delta=0.014 in a constructive refinement, and Gowers calling it Annals-of-Mathematics quality. The bigger signal isn't the proof. It's Shankar's CoT observation: most of the model's reasoning attempted counterexamples to the conjecture, not validations of it. That's calibrated contrarianism — a scorable behavioral property and the math-grounded analogue to sycophancy detection. Verifier-rich domains are where autonomous AI lands first; counterexample-seeking is how we'll measure whether reasoning is real or performative.

# tags
openaiai-for-scienceverifier-bottleneckagentic-ai-viabilityfrontier-modelsautomated-researchevalrigrecursive-self-improvementcapability-overhangharness-as-moatresearch-methodologyai-economicsai-labor-displacementai-1.0-defensibility
◆ entities
OpenAIErdosTim GowersNoga AlonArul ShankarJacob TsimermanThomas BloomWill SawinPrincetonEvalRigPickRig
→ threads
verifier-bottleneckagentic-ai-viabilityai-1.0-defensibility
⟷ links
2026-04-14-12026-03-27-22026-03-13-12026-03-21-22026-03-31-m22026-04-04-32026-04-14-22026-04-17-32026-04-25-32026-04-30-12026-05-01-22026-05-11-32026-05-13-2
permalink