AI agents are fast, loose and out of control, MIT study finds
February 20, 2026
•ZDNet_AI•Moderate impact[key points]
Key points
- •An MIT study reveals that AI agents are fast, loose, and out of control.
- •Agentic AI technology is marked by a lack of disclosure about risks.
- •AI developers need to step up and take responsibility, improving transparency and control of these systems.
Impacted Jobs
View profile →
View profile →
Coming soon
Full job impact analysis
The premium watch subscription (detailed per-job impact, extended access) is coming soon. In the meantime, enjoy the key takeaways and highlighted roles on each article.