Researchers Debate: Today's Models Stay Brittle Outside Spoon-Fed Zones
anshulkundaje · x · 2026-09-16
Genomics researcher Anshul Kundaje debates the limits of the current AI paradigm: increasingly powerful models remain "utterly daft" in zones they weren't spoon-fed, making training fragile in unpredictable ways. He adds he's unconvinced that truly intelligent models with real understanding would be so obliviously unintentionally misaligned.
Related event: Stanford Scholars Warn Stronger Models Are More Opaque and Fragile(2 posts)→
More from AGI Musings
- What do we call the 'it was ever thus' rhetoric used to dismiss AI concerns? — jjvincent · 2026-09-16
- AI's hardest problems need democratic deliberation — and independent experts — RishiBommasani · 2026-09-16
- Pedro Domingos: AI is anti-moat, dissolving switching costs that protect IT providers — pmddomingos · 2026-09-16
- Founder pushes back on Anthropic CEO's runaway-AI warnings on NDTV Profit — angadc · 2026-09-16
- AI Safety comms debate: punchy messaging wins short-term but erodes community epistemics — NathanpmYoung · 2026-09-16
- AI sentience debate reignites as critics call TV claims 'made-up numbers' — suchenzang · 2026-09-16