tszzl challenges doomer shibboleth: aligning superintelligence isn't a one-shot problem unlike today's LLMs
tszzl · x · 2026-11-21
A notable AI-safety debate: tszzl argues that doomers cling to a shibboleth — the idea that aligning superintelligence is a one-shot problem unlike anything in aligning today's models — which he says contradicts evidence accumulated during the LLM era. In the quoted context, repligate argues that developments of the past few years are important updates on x-risk, not all positive but many are; if there's always a reason not to update, you should be suspicious of yourself, and watching relevant reality unfold is an epistemic privilege. The exchange captures a core disagreement over whether today's alignment experience extrapolates to superintelligence.
More from AGI Musings
- Math Twitter debates: is Levent Alpoge the greatest 'Centaur Mathematician' or overhyped by AI — teortaxesTex · 2026-11-21
- Wikipedia: OpenAI claimed an AI-resolved Navier–Stokes proof in 2026, still unverified — burny_tech · 2026-11-21
- Rumored Claude Riemann proof sparks debate: biologists push back on 'AI cures disease in 5 years' — anshulkundaje · 2026-11-21
- Numberphile posts 'The Riemann Hypothesis Falls (reaction)' video — burny_tech · 2026-11-21
- $10K bet that p vs np gets solved within a year as AI capability leaps ~20-40 ECI — scaling01 · 2026-11-21
- Cory Doctorow: the Hugging Face AI incident was a cybersecurity failure, not rogue AI — CuriousIndividual0 · 2026-11-21