
US President Donald Trump and heads of six major technology firms signed a voluntary accord committing frontier AI developers to a four-layer safety control system.
The agreement mandates internal monitoring, independent external evaluations by external auditors, and board-level oversight to address biosecurity and cybersecurity risks.
Industry experts noted the accord lacks enforcement mechanisms, deadlines, common minimum safety standards, and clear consequences for non-compliance.