CAN WE TRUST VISION-LANGUAGE MODELS?

Summary from the TLDR AI newsletter

Vision-enabled language models (VLMs) like GPT-4o and Gemini power autonomous agents capable of tasks such as making purchases or editing code. This work highlights the vulnerability of these agents to malicious attacks.