#MathProofs
“Toxic waste.”

Fields Medal-winning mathematician Terence Tao reflects on the hundreds of “terrible proofs” released by #OpenAI earlier this month. #MathProofs
Terence Tao: AI turns beautiful mathematics into ‘toxic w...
The top mathematician says Silicon Valley’s rush to solve long-investigated conundrums excludes ‘elegant solutions’ and reduces maths to raw training material
observer.co.uk
October 26, 2026 at 7:22 PM
OpenAI’s new AI agents are solving thousands of math puzzles in parallel— from Navier‑Stokes to Millennium‑prize style proofs, even checking Lean verifications. Imagine a swarm crunching problems we thought needed human insight. Dive in! #AIagents #MathProofs #ParallelProcessing

🔗
September 30, 2026 at 2:39 PM
AI aces basic math but struggles with complex proofs like the US Math Olympiad. Research reveals even 'reasoning' models lack deep understanding for competition challenges, often confidently wrong. Pattern matching isn't true proof. #AI #LLM #MathProofs
April 25, 2025 at 11:05 PM
DeepSeekMath‑V2 is an open‑source beast that writes and checks its own math proofs, using an internal critique loop and low test‑time compute. Could it be the pop‑off point for the US AI bubble? Dive in to see how it stacks against OpenAI and DeepMind. #DeepSeekMathV2 #OpenSourceAI #MathProofs

🔗
November 29, 2025 at 5:33 PM
Overview: Hacker News discusses formalizing Fermat's Last Theorem using the Lean theorem prover. Key themes include the structure of Focused Research Organizations (FROs) and the potential role of AI in rigorous mathematical proof verification. #MathProofs 1/6
August 5, 2025 at 4:00 AM
HN discussed "proofs without words" - visual math proofs w/o formal text. Users debated their appeal vs. limitations, rigor concerns, and subjectivity. Some find them intuitive, others misleading. Explores resources & examples. #MathProofs 1/5
June 19, 2025 at 2:00 PM