US Senators Seek Clarification from OpenAI on AI Safety and Employment Practices

The letter is fueled by the recent reports that question OpenAI’s commitment to its goals of safe and responsible AI development.
US Senators Seek Clarification from OpenAI on AI Safety and Employment Practices
Representative image by freepik.
Published on

OpenAI CEO Sam Altman has received a letter from Senate Democrats demanding clarity on the AI safety and employment practices of the company. Signed by four senate Democrats — Brian Schatz, Ben Ray Luján, Peter Welch, and Mark R. Warner — and Junior Senator Angus S. King, the letter is fueled by recent reports that question OpenAI’s commitment to its goals of safe and responsible AI development.

“Given OpenAI’s position as a leading AI company, it is important that the public can trust in the safety and security of its systems,” the senators wrote. “This includes the integrity of the company’s governance structure and safety testing, its employment practices, its fidelity to its public promises and mission, and its cybersecurity policies.”

The lawmakers have asked OpenAI to furnish detailed information on the following key areas by August 13, 2024:

  • OpenAI’s commitment to dedicating 20% of its computing resources to AI safety research.

  • The company’s stance on non-disparagement agreements for current and former employees.

  • Procedures for employees to raise cybersecurity and safety concerns.

  • Security protocols to prevent theft of AI models, research, or intellectual property.

  • OpenAI’s adherence to its own Supplier Code of Conduct regarding non-retaliation policies and whistleblower channels.

  • Plans for independent expert testing and assessment of OpenAI’s systems pre-release.

  • Commitment to making future foundation models available to US government agencies for pre-deployment testing.

  • Post-release monitoring practices and learnings from deployed models.

  • Plans for public release of retrospective impact assessments on deployed models.

  • Documentation on meeting voluntary safety and security commitments to the Biden-Harris administration.

“National and economic security are among the most important responsibilities of the United States government, and unsecure or otherwise vulnerable AI systems are not acceptable,” the letter states.

Related Stories

No stories found.
CDO Magazine
www.cdomagazine.tech