Our summary
The author explores how to use Jev, an AI tool that chooses from options rather than writing an answer, to improve computer security. Instead of having the software investigate freely, security teams could use it to score and rank suspicious events so humans know what to look at first.
The software looks at large amounts of computer activity and assigns a threat score to different users or programs. It can also check if an automatic security alert actually matches the specific behavior its rule claims to find, helping to organize long lists of events.
This method is meant for security professionals who need to organize their work, not to replace human judgment or existing security rules. The author is still testing these ideas, so the suggested benefits for saving time and money have not been proven in a finished project.
Key takeaways
- Score users and computer programs based on how suspicious their recent activity looks.
- Use the AI tool to check if an automatic security alert matches its claimed behavior.
- Shorten investigation times by having the software highlight only the most relevant events for human review.