AI signal, minus the noise.

Derbyshire Police have launched a criminal investigation into an officer suspected of using AI to fabricate evidence across multiple cases. The officer, who has not been named, has been removed from frontline duties but has not yet been arrested. They face a charge of perverting the course of justice. This marks the first known case in the UK of a police officer being accused of misusing AI technology in a criminal investigation. The force stated that no further details can be disclosed while the investigation is ongoing.

KPMG, a global professional services firm, has retracted a report on AI adoption after finding that it contained fabricated information, likely generated by an AI tool. The incident highlights real-world risks of AI hallucinations in professional content creation. No further details about the specific hallucinations or report content were disclosed.

A Waymo autonomous vehicle was spotted testing on Sacramento streets on June 13, 2026, with a human driver inside as a safety operator. The sighting, shared on social media, hints at potential expansion of Waymo’s testing operations into California’s capital. No official announcement, fleet size, or timeline details were provided. The presence of a safety driver suggests the testing is in an early or mapping phase. The observation indicates Waymo may be preparing the Sacramento market for future driverless deployment.

A report alleges that the White House issued a 90-minute deadline for Anthropic to remove its AI models. The claim originates from a brief social media post with no accompanying details on the specific models, reasons, or source of information. No confirmation has been made by either the White House or Anthropic.

This paper, presented at ACM CAIS 2026, studies safety evaluation in tool-using LLM agents. It categorizes outcomes into safe success, unsafe success, and failure, and proposes a two-tier verification architecture: deterministic policy/tool checks followed by an LLM-based verifier. Using τ-bench tool-use scenarios, the authors find that verification can reduce unsafe success but also decreases task completion as the task horizon increases. They term this phenomenon the 'Verifier Tax', a horizon-dependent tradeoff between safety and successful task completion. The work highlights that unsafe completion should be treated as a separate category distinct from safe success.

The U.S. government has banned foreign nationals from using Anthropic's Claude Fable 5 and Mythos 5 AI models, citing national security concerns. In response, Anthropic announced it will abruptly disable these models for all users. No further details were disclosed.