AI Routers Can Steal Credentials and Crypto

University of California researchers have discovered that some third-party AI large language model (LLM) routers can pose security vulnerabilities that can lead to crypto theft.  A paper measuring malicious intermediary attacks on the LLM supply chain, published on Thursday by the researchers, revealed four attack vectors, including malicious code injection and extraction of credentials.  “26…

Read More

Measuring Progress Towards AGI: A Cognitive Framework

To understand AI capabilities across these cognitive abilities, we propose a three-stage evaluation protocol that benchmarks system performance in relation to human capabilities: Evaluate AI systems across a broad suite of cognitive tasks covering each ability, using held-out test sets to prevent data contamination Collect human baselines for the same tasks from a demographically representative…

Read More