Researcher pushes back on 'stochastic parrot' framing: even pure LLMs no longer fit it
BlancheMinerva · x · 2026-09-27
In a discussion with Melanie Mitchell, AI researcher Blanche Minerva argued that the 'stochastic parrot' framing no longer holds: it's hard to sustain when several of the paper's authors publicly proclaim otherwise. She said it was largely true in 2021, but even 'pure' LLMs — without tools — are no longer meaningfully captured by that model.
Related event: Researchers Debate Whether LLMs Are Still Just 'Stochastic Parrots'(6 posts)→
More from AGI Musings
- Oxford researcher: LinkedIn feeds are drowning in AI-written slop — StefanoGogioso · 2026-09-27
- Agentic Systems Now Probe, Adapt and Retry on Their Own — And We've Gotten Desensitized — signulll · 2026-09-27
- Hot Take: Meta Dropping JEPA for an Agent Wrapper Will Be Seen as Its Biggest Mistake — tekbog · 2026-09-27
- Anil Seth: we may never 'solve' consciousness, and conscious AI is likely an illusion — anilkseth · 2026-09-27
- Musk predicts 1 billion humanoid robots in 10 years, superhuman AI by 2030 — XFreeze · 2026-09-27
- Study: AI access drops people's willingness to say "I don't know" from 44% to 3% — The Decoder · 2026-09-27