Anthropic Drops AI Safety Pauses Amidst Competition, Regulatory Gaps

• Anthropic has released a new version of its Responsible Scaling Policy (RSP), weakening a cornerstone safety pledge.
• The company will no longer pause or delay advanced model training even if abilities exceed internal safety controls.
• Anthropic cited "heightened competition" and a lack of effective AI regulation as reasons for the policy shift.
• Chief Science Officer Jared Kaplan stated the original policy fell behind the pace of technological change and unilateral commitments were impractical.
• The policy revision comes amidst tensions with the US Department of Defense, though Anthropic claims the timing is unrelated.