MIT launches the LLM Election Observatory, a dashboard tracking how nearly a dozen AI models tailor responses to political queries during the 2026 US midterms (Tiffany Hsu/New York Times)

MIT launched the LLM Election Observatory, a dashboard that measures how roughly a dozen AI models answer political questions ahead of the 2026 U.S. midterm elections. Anthropic’s threat-intelligence report details how it intervened to stop several misuse attempts of its Claude model, including cyber-attacks, influence and surveillance operations, weapons-related research, scams, illicit distillation and other malicious activities, shutting down work whose intent could not be verified.