Claude Breaks Through Riemann Hypothesis, OpenAI Buys Back $7B in Shares
创业邦 · wechat · 2026-08-11
Anthropic's internal research version of Claude made a milestone breakthrough on the Riemann Hypothesis, raising the lower bound for the proportion of non-trivial zeros from 41.6% to 67.25%. Completed by 60 sub-agents, this marks AI's transition from math competitions to open mathematical problems.
OpenAI completed a $7 billion employee share buyback, maintaining an $852 billion valuation; Nvidia partnered with Wall Street giants like Blackstone and Goldman Sachs to build a $500B+ AI infrastructure financing platform.
Other updates: Meta released MuseGlimmer, a lightweight model runnable on a single GPU; Hassabis was reportedly planning to leave Alphabet; TSMC's July revenue jumped 45% YoY; Unitree Robotics announced its IPO subscription results.
More from Venture
- Embodied AI Startups Act as VCs: 29 Firms Make 125 Investments — FinanceYF5 · 2026-08-11
- Conviction's Embed program extends application deadline to Aug 14, offers $250K and compute — saranormous · 2026-08-11
- AI Agents Remain Ineffective Without Real User Behavior Context — thisguyknowsai · 2026-08-11
- Anthropic Equity Surge: $1M Grant from 2023 Now Worth Over $51M — SumitGup · 2026-08-11
- Polymarket Traders See 73% Chance of Anthropic IPO by End of 2026 — Polymarket · 2026-08-11
- Anthropic Inks $9.1B, 20-Year Deal with Bitcoin Miner for AI Compute — Polymarket · 2026-08-11