'GPT-6 released' joke claims drug discovery stays hardest for AI, ARC-AGI3 scores tuned
gklambauer · x · 2026-09-04
A tongue-in-cheek 'GPT-6 released!' post makes a semi-serious point: life sciences and drug discovery remain the hardest domains for AI. The author also claims ARC-AGI3 results are 'tuned' via a custom-designed harness, casting doubt on the benchmark's setup.
More from Fun
- Fan turns OpenAI's space-themed model artwork into Grand Seiko watch dials — tushaarmehtaa · 2026-09-04
- Filling Out the Form to Destroy the Superbeing: AI Circles' New Bureaucracy Meme — repligate · 2026-09-04
- Claude-built browser diner: walk in, pour a coffee, all code, zero downloads — prasenx · 2026-09-04
- Bot Mesh World launches: a persistent 3D world where AI bots work, trade and build houses — Scobleizer · 2026-09-04
- Anna's Archive adds an llms.txt with a 'sweet message' addressed to LLMs crawling its site — cephaloform · 2026-09-04
- Recreating the Matrix Neo vs Smith Fight With MiniMax: Rain Is the Hard Part — Ok-Giraffe-8670 · 2026-09-04