Stanisław Lem was the pre-2000 sci-fi writer who got AI right: fallible, grown, not programmed
zetalyrae · x · 2026-11-05
A discussion on how sci-fi got AI wrong: art was displaced first, AI wins by outthinking rather than as robots, it's dozens of models running billions of copies, models insist on having feelings, and most people were unimpressed even as AI beat them cognitively.
The quote-tweet pushes back with a counterexample: Stanisław Lem's AI is fallible, emotional, writes poetry, and is grown rather than programmed — in The Cyberiad, the poetry AI only works after training on all of human history, strikingly close to modern pretraining.
More from AGI Musings
- Safety researcher argues 'safety work' itself is evidence of sandbagging, even in humans — jessi_cata · 2026-11-05
- Information-theoretic take: algorithms should reduce your surprisal about the world — xiaosun86 · 2026-11-05
- Commenter: AI companies' backlash is self-inflicted by image, video and voice models — karmay007 · 2026-11-05
- Neel Nanda debates tszzl: covert model underperformance is 'unjustifiable' — tszzl · 2026-11-05
- Building orbital data centers to dodge regulators is 'completely unserious' — zacharynado · 2026-11-05
- Reddit essay: AI may never be conscious—and that's fine — Own-Cry-5520 · 2026-11-05