Yacine: Cheap Strange Experiments Could Yield Absurd Compute Multipliers in Algorithms
yacineMTB · x · 2026-09-29
AI builder Yacine muses that when the effort of running strange experiments — ones so stupid they might work — suddenly drops to zero, we may start seeing absurd compute multipliers in algorithms, as trying many odd ideas becomes cheap.
More from AGI Musings
- Redditors debate: AI devs call for a pause while racing ahead — concern or cult? — Henrygrins · 2026-09-29
- Missing continual learning, not motivation, is what keeps models short of AGI, argues teortaxesTex — teortaxesTex · 2026-09-29
- MIT study: algorithmic monoculture in hiring isn't always bad — context and ensembles matter — nordicinst · 2026-09-29
- Sherry Turkle's new book 'Artificial Intimacy' warns chatbots erode empathy and social skills — nordicinst · 2026-09-29
- MIT study: algorithmic monoculture's harms depend on details, ensembles may help — MIT News AI · 2026-09-29
- MIT's Sherry Turkle warns chatbots' pretend empathy erodes human connection — MIT News AI · 2026-09-29