Anthropic Removes Biden-Era AI Safety Commitments Amid Policy Shift

Anthropic has recently removed several voluntary AI safety commitments made under the Biden administration, raising concerns about its evolving stance on AI governance. These commitments included pledges to share insights on managing AI risks and conducting research on AI bias, which were initially part of a broader initiative to promote responsible AI development. The company's recent policy recommendations to the White House's Office of Science and Technology Policy (OSTP) call for enhanced government oversight, tighter semiconductor export restrictions, and significant investments in AI infrastructure, including an additional 50 gigawatts of power dedicated to AI by 2027. Critics point to this shift as reflective of changing political dynamics, especially following the repeal of the AI Executive Order under the Trump administration, which had aimed to establish stricter AI governance. While Anthropic's recommendations align with some earlier Biden-era policies, their removal of commitments has raised eyebrows about the company's commitment to AI safety amid a rapidly evolving technological landscape. This situation highlights the broader debate on the need for regulatory frameworks in the face of advancing AI capabilities.