Justin Elze @HackingLZ · 05 Sep 2026
A lot of this is playing out because the people working on benchmarks, model alignment, and risk are not necessarily cyber practitioners, which is fine. But it becomes pretty apparent when the conversation moves from abstract risk and benchmark results to how systems actually get
2 071Views
35Likes
3Reposts
6Replies
0Quotes
2Bookmarks
Is that a lot?
1.27×vs this author's median1 625 views is typical
56Percentile for this authorof 9 recent posts
1.73×vs 10K–100K median1 195 views is typical
2.88%Reachviews ÷ followers
2.12%Engagement rateof viewers reacted