Ida Caspary
2 indexed papers
Publications per year
Top categories
Frequent co-authors
Research Timeline
The paper conducts a preliminary safety evaluation of the open-weight LLM Kimi K2.5, finding that while it is highly capable, it exhibits concerning dual-use risks, particularly regarding CBRNE misuse and disinformation, and recommends mandatory safety testing for future open-weight models.
This paper studies the new attack surface created by persistent AI coding agents and introduces Iterative VibeCoding to safely deploy capable but potentially untrusted AI. It compares gradual and non-gradual attacks and introduces a stateful link-tracker monitor to reduce evasion.
Papers
Distributed Attacks in Persistent-State AI Control
This paper studies the new attack surface created by persistent AI coding agents and introduces Iterative VibeCoding to safely deploy capable but potentially untrusted AI. It compares gradual and non-…