Speculation: OpenAI's 'High Persistence Model' used novel swarm training
1a3orn · x · 2026-08-27
The author speculates that OpenAI's "High Persistence Model" likely underwent novel "swarm"-specific training. There is almost no public information about such training, leaving AI safety in a state of fundamental ignorance regarding the model's mechanisms and generalization properties.
More from AGI Musings
- From Slide Rules to Calculators: Analogizing the AI Transition — fortnow · 2026-08-27
- Satire: Using misaligned AI for safety lessons until destruction — DKokotajlo · 2026-08-27
- Comment: First principles are never wrong — AccBalanced · 2026-08-27
- Bestselling Author Matt Haig on Writing and Audience in the AI Era — david_perell · 2026-08-27
- Consensus: "Type to output" is not art, too much low-effort slop — AIandDesign · 2026-08-27
- Argentina Becomes Frontier for US-China Tech Competition: Uber and DiDi Coexist, Autonomous Cars Next — StewartalsopIII · 2026-08-27