AI practitioner calls ARC-AGI benchmark a 'serious taste issue', sparking debate
HanchungLee · x · 2026-09-07
X user Hanchung Lee criticized the use of ARC-AGI as a model capability benchmark, calling it a "serious taste issue." The post sparked a debate over whether AGI exists at all, with one commenter saying that despite admiring Jensen Huang, he's "simply wrong this time."
More from Models
- Gary Marcus cites hands-on review: Astra is the best planner but a terrible coder — GaryMarcus · 2026-09-07
- GPT-6 Astra rumor: 1.2T active params, trained on Abilene's 100-150k GPUs — teortaxesTex · 2026-09-07
- Meme: Google may deem Gemini 4 Pro not worth it and pivot back to Flash models — Able-Line2683 · 2026-09-07
- Astra robot-control demos pile up; speculation OpenAI is building its own Gemini Robotics rival — CyberRobooo · 2026-09-07
- Professor says Astra drafted five NIH R01 grant proposals in an hour for ~$20 of compute — mmbronstein · 2026-09-07
- GPT-6 "Astra" shows zero-shot robot arm sorting; OpenAI robotics rumors swirl — CyberRobooo · 2026-09-07