NYU professor defends Cyc's place in AI history as LLMs finally crack common sense
ipeirotis · x · 2026-09-29
NYU professor Panos Ipeirotis defended Cyc's place in AI history: it was one of the earliest serious attempts at common sense, and dismissing it ignores decades of AI research. With LLMs and foundation models, systems can finally draw on broad everyday context with a breadth and flexibility earlier approaches never achieved. He also rejects the metaphysical "understanding vs. simulating understanding" debate—if the two make no experimentally distinguishable predictions, there's nothing to adjudicate.
More from AGI Musings
- "ChatGPT will cut plumbing demand" claim sparks mockery: critic says poster doesn't know what plumbers do — ctjlewis · 2026-09-29
- After ~20 papers, economists' verdict: AI hasn't hit aggregate labor market yet, but junior white-collar hiring may already be sliding — soumitrashukla9 · 2026-09-29
- Ben Lorica: the AI data problem has moved downstream — from raw material to usable-data labor and licensed access — bigdata · 2026-09-29
- "You are but vessels for LLMs": dev mocks prompt-replaces-art takes — threepointone · 2026-09-29
- Blogger predicts AI with CEO-level skills will automate company formation within a year — _AustinCalvert_ · 2026-09-29
- Redditor calls out AI companies' 'too powerful to release' narrative as PR spin — vinmi · 2026-09-29