Interesting part is the gap between beliefs and prices => if the investors expect a US debt crisis we expect it to show up somewhere in the term premium real yields dollar or inflation expectations ; those signals can remain muted for a long time
595 cumulative year of litigation is probably most important statistic here => even when governments eventually win the process itself can be delay policy for years
The most significant part isn't the zero-day but its the model ability to autonomously plan adapt and chain multiple exploits toward a long term goal => that raises the bar for AI security evaluations and defensive tooling alike