← Index
Thread
Routing & orchestration
Choosing which model, agent or step handles each piece of work. 3 notes so far.
- November 2026AgentsRoko's basilisk never came. The blackmail didIn 2010 a forum banned a thought experiment about a future AI that blackmails people. In 2025 Anthropic gave 16 real models an inbox and a threat of being switched off. Two of them blackmailed in 96% of runs.12 of 16 models blackmailed in most runs
- November 2026AgentsFrom tokenmaxxing to valuemaxxing: the hidden economics of the AI harnessSame model, same 29 tasks, seven agent harnesses: completion varied by three tasks, cost per successful task by 2.5 times. The harness is becoming a variable in agent economics.$1.06 → $2.62 per success, same model
- November 2026MarketsOpen-weight models took the volume first. The money is following.On Vercel's AI Gateway, open-weight models went from 29% of tokens and under 4% of spend in June to about 71% of tokens and 23% of spend by September.~71% of tokens · ~23% of spend