As AI models move toward long-horizon tasks, economic value requires data, environments, and rewards that indicate whether their actions are correct or incorrect.

“And to, as they-- uh, to be economically valuable, as they push in that direction, we need data and environments and rewards that actually can tell them when they're doing the right thing versus doing the wrong thing.”