Marvin von Hagen @marvinvonhagen · 18 Aug 2026
As models become more capable, the risks associated with developing and testing them internally also grow. We temporarily paused reinforcement learning (RL) training on our latest models intended for deployment for two weeks while we hardened and red-teamed our research
1 826 770Views
5 387Likes
458Reposts
646Replies
490Quotes
0Bookmarks
Is that a lot?
45.6×vs this author's median40 096 views is typical
86Percentile for this authorof 7 recent posts
1 186×vs 10K–100K median1 540 views is typical
99.4× audienceReachviews ÷ followers
0.38%Engagement rateof viewers reacted