In other AI news, prompt injections are here to stay based on this new paper from Abdelnabi and Bagdasarian. "an adversary can always construct a context under which a blocked flow appears legitimate, or a defender who tightens norms will block genuinely legitimate flows." arx…
Who's Who of AI
Mor Naaman
Cornell Tech professor (information science, AI-mediated Communication, trustworthiness of our information ecosystem). New York City. Taller in person. Opinions my own.
What they're sharing
AI Agents May Always Fall for Prompt Injections arxiv.org
youtube.com
Articles & links
Hilarious (and a bit sad) that Beato checks AI Slop videos using... an AI chatbot (and did not even check that the results were coherent, which they weren't). Also, the platforms should start doing something about AI Slop. www.youtube.com/watch?v=PUSY...
Their own posts
Recent commentary
Wow this is some wild hallucination, Google! (Found on the Discover feed).
Their network
In Mor Naaman's orbit
Center = Mor Naaman. Left = members they follow (green edges). Right = members who follow them (blue edges). Top = mutual follows (orange edges, slightly larger). Drag any node to reposition; click to open that profile.