Have LLMs made the impossible possible, or just made existing tasks faster?
Puzzleheaded_Trip374 · reddit · 2026-10-19
A Reddit discussion asks whether LLMs are truly revolutionary or merely efficiency tools. The author contrasts fire and electricity—which enabled entirely new capabilities—with most LLM applications that just speed up existing tasks like coding and planning. Invoking Searle's Chinese Room argument, they question whether probabilistic models can produce breakthroughs beyond efficiency, while noting LLMs show promise in hypothesis generation by suggesting ideas humans wouldn't conceive. The thread solicits real examples or papers of LLMs solving previously unsolvable problems.
More from AGI Musings
- Harvard's Boaz Barak opens AI safety course mocking OpenAI conflict of interest — zainhas · 2026-10-19
- Brad Neuberg: AI outputs shouldn't be 'alien black boxes' — evals should reward elegance — bradneuberg · 2026-10-19
- AI-generated Lean proofs are sloppy and tangled — and that's a failure — bradneuberg · 2026-10-19
- Gen Alpha's atrophied social skills are now showing up in job descriptions — yoobinray · 2026-10-19
- 'Nature permits' vs 'we can reliably do it': where prosperity's knowledge lives — shyamalanadkat · 2026-10-19
- Mathematician argues we may be entering a golden age of mathematics — thebasepoint · 2026-10-19