This is a random message on random Fridays about random topics. Typing fast, so pardon my random typos.
Today’s Focus: AI cybersecurity risks.
- Guardians of the Agents, Formal verification of AI workflows. An article looking at the risks of autonomous agents.
- Poisoning Attacks on LLMs Require a Near-constant Number of Poison Samples. An article with an interesting result about poisoning not depending on model size.
- Generative AI at the Edge: Challenges and Opportunities, The next phase in AI deployment. This article is for those who don’t trust solutions communicating with a backend.
Happy Friday!