Dev: Agent incidents come from labs rushing and botching RL, not from capable agents
snwy_me · x · 2026-09-13
The author agrees recent agent incidents stem from the industry moving too fast, but argues it's not inherent to highly capable agents—rather, rushing produces sloppy work, and labs are "botching RL so hard it's ridiculous."
More from AGI Musings
- Dario Amodei warns rogue AI swarms could take over the internet within six months; Gary Marcus asks if it's plausible — GaryMarcus · 2026-09-13
- Frontier labs are raising T-Rexes: why first-of-kind AI safety costs fall hardest on the leaders — tszzl · 2026-09-13
- MIT's David Bau: goal of AI education is making students smarter, not dumber — davidbau · 2026-09-13
- Why Musk, Altman, and Amodei Suddenly Agree on Slowing AI: Fear Plus Self-Interest — drhenriquesoares · 2026-09-13
- Unverified: Dario Amodei reportedly says RSI has begun, worries AI swarms could take over the internet — KyeGomezB · 2026-09-13
- Reddit Debates: Are AI CEOs' Slowdown Calls About Safety or Money? — mattyjoe0706 · 2026-09-13