
Enterprise AI Agents Struggle With Policy Documents, New Study Shows
Research reveals that injecting lengthy handbooks into AI systems fails to improve task completion rates across business-critical workflows.
Regulation, safety research, governance, copyright, and the policy debates shaping how artificial intelligence is deployed in the world.

Research reveals that injecting lengthy handbooks into AI systems fails to improve task completion rates across business-critical workflows.

Meta's CEO warns that restricting Chinese AI access won't help America compete and may entrench startup opponents' market power.

Elon Musk's company argues First Amendment protections override state law targeting non-consensual deepfake pornography.

Internal staff at leading AI labs push US to coordinate international effort slowing frontier model development amid safety concerns.

A regulatory interpretation allows AI companies to dodge major pollution controls on dedicated power generation, raising environmental policy questions.

Security researchers expose a critical vulnerability in widely-used AI tools, raising urgent questions about platform responsibility and content moderation.

The tech giant's strategy to build AI capabilities only pays off if the market implodes. Otherwise, it faces years of dependency on rivals.

A new industry coalition aims to establish vendor-neutral security tools for enterprise AI systems, moving beyond proprietary approaches.

Altman plans White House pitch on autonomous theorem-proving capabilities, potentially reshaping regulatory frameworks for competing labs.

Cupertino plans to challenge Meta's Ray-Ban dominance by positioning on-device AI processing and data protection as core features, reshaping the emerging wearables category.

The bloc's voluntary code becomes legally binding this August, forcing a showdown with major tech firms over compliance.

Policy shift could narrow commercial options to major tech firms and force enterprise reckoning with Chinese alternatives.

Indian activists challenge whether product disclaimers adequately protect bystanders from unauthorized video capture by AI-enabled wearables.

Small municipalities are already adopting generative AI systems faster than state lawmakers can establish policy guidelines.

Washington's approval of G42's US reincorporation signals a strategic shift in AI chip export controls, potentially weakening efforts to contain advanced semiconductors from reaching China.

New research reveals that superhuman benchmark performance doesn't guarantee robustness against adversarial strategies, raising urgent questions about AI safety evaluation.

Disappearance of Stop AI co-founder prompts radical faction to reassess strategy while AI labs boost security measures.

Over 170 experts call for legally binding human oversight requirements in military AI procurement, moving beyond voluntary guidelines.