badlogic: 'training data/RL envs is all you need' for shaping model behavior
badlogicgames · x · 2026-10-16
@badlogicgames (Mario Zechner) asserts in a discussion with @willcb that "training data/RL envs is all you need" — the key to shaping model behavior lies in training data and reinforcement learning environments rather than other techniques.
willcb responds that open specifics aside, nothing seems particularly daunting, and sees a nearly practical design space for higher-level behavior shaping. The one-liner captures the current consensus around improving models and agents via RL environments.
Related event: Researchers Debate: Training Data and RL Envs Are All You Need(2 posts)→
More from AGI Musings
- An Atheist Defends the Concept of Soul, Grounding AI Moral Status in Inner Experience — esrtweet · 2026-10-16
- VC warns a "digital aristocracy" of LLM-fluent workers is emerging — StewartalsopIII · 2026-10-16
- DHH: Starship, Starlink and Grok should have been Europe's Sputnik moments — georgemillo · 2026-10-16
- Hinton hits back at Jensen Huang: 'He has good reason to say AI isn't dangerous — he sells Nvidia chips' — BlackHC · 2026-10-16
- Yudkowsky blasts new interpretability finding as primitive; critic pushes back on doomer logic — AlexTensor · 2026-10-16
- Beff Jezos declares 'civilizational inflection point' in the acceleration of technological progress — beffjezos · 2026-10-16