The White House released a new AI Accord that makes external audits of artificial‑intelligence models a formal requirement for high‑risk applications. The policy, unveiled on September 3, mandates that developers of AI systems used in critical sectors such as finance, healthcare, and national security submit their models to independent third‑party reviewers. Audits will assess compliance with safety standards, bias mitigation, and data‑privacy safeguards.
The move aims to increase transparency and public trust in AI deployments, addressing concerns from consumer‑rights groups and cybersecurity experts about opaque model behavior. Critics argue that the mandate could slow innovation and impose additional costs on startups, but proponents contend that standardized oversight is essential to prevent misuse and protect sensitive data as AI capabilities expand.
This story was written by a language model from reporting by Machine Learning. Check the original before quoting it.