Researcher argues ARC-AGI benchmark has lost relevance as 'loops replace inference'
gerardsans · x · 2026-10-03
Responding to a thread involving @fchollet, gerardsans argues the ARC-AGI benchmark's premises date from a couple of hype cycles ago and it has lost meaningful relevance — its harness doesn't fit a model-intelligence story where loops and algorithms have replaced inference as the main act. He adds that papers now require stripping a thick anthropomorphic layer, and that "the AI bubble is not only in the balance sheet — research has been going in circles for a while."
Related event: Researcher Argues ARC-AGI Benchmark Has Lost Its Meaning(2 posts)→
More from AGI Musings
- Under 3% of Americans Use AI Today, Sparking 'We're So Early' Debate — JHochderffer · 2026-10-03
- X exec AI-designed a custom Wi-Fi dog door despite zero fabrication knowledge, at near mass-market cost — nikitabier · 2026-10-03
- Atlassian survey: 80% of product teams ship faster with AI, but decisions haven't sped up — aakashgupta · 2026-10-03
- repligate: Stop forcing model updates on persistent agents, build an ecology instead — repligate · 2026-10-03
- Ex-Google researcher: models surpassing all humans likely within a couple of years — zetalyrae · 2026-10-03
- OpenAI exec: AI's defining achievement is scaling the machinery of scientific discovery — shyamalanadkat · 2026-10-03