Claude runs 53 hours unsupervised, beats human designs on 7 of 8 quantum problems
FinanceYF5 · x · 2026-10-26
In an unsupervised closed loop, Claude ran for 53 hours on eight quantum computing problems and outperformed the best human designs on seven. Takeaway: AI can do real research wherever automated scoring exists — making high-quality graders/rubrics more valuable than smarter models.
More from Models
- Bindu Reddy: US open source models are 'insanely behind' Chinese models — bindureddy · 2026-10-26
- Gemini 4 Flash spotted in Antigravity: internal build gemini-4-cp with 200k context — koltregaskes · 2026-10-26
- Depth-Grown and Looped LMs Learn In-Context Learning Earlier Than Dense Models — omarsar0 · 2026-10-26
- Current LLMs Are Horrible at Type Design in Dependent Type Languages — hargup13 · 2026-10-26
- Chart: Claude Opus 5.5 hits 34.4% of token spend within two weeks, chasing Opus 4.6's peak — scaling01 · 2026-10-26
- Developer: Claude Is Bad at Security Audits — letandrewcook · 2026-10-26