Four report cards in one week: Step 5 Preview benchmarks, Jev's confidence blind spot, and what California's AI order actually requires

每个数字各在测什么：一周内的四份成绩单

StepFun's Step 5 Preview posted four types of scores, but its self-reported Terminal-Bench result isn't on the public leaderboard yet, and weight license terms remain undisclosed. An independent dev tested TypeSafe's Jev classifier on 111 public hard questions: Jev's average confidence was 0.690 when wrong and 0.684 when right—confidence doesn't separate correct from incorrect. California's AI executive order only directs state agencies to draft legislative proposals; the only active law binding companies is 2025's SB 53.