Senate Democrats and one independent lawmaker have sent a letter to OpenAI
CEO Sam Altman, requesting information on the company’s safety standards and employment practices toward whistleblowers. The letter also seeks a commitment to allow U.S. government agencies to test and review OpenAI’s foundation models before deployment.
Points
- Senate Democrats and one independent send letter to OpenAI CEO Sam Altman
- Requests information on safety standards and whistleblower practices
- Seeks commitment to allow U.S. government agencies to test foundation models pre-deployment
- Concerns over AI safety and potential malicious use
- Increased regulatory scrutiny of OpenAI and the AI sector
Senate Democrats and one independent lawmaker have addressed a letter to OpenAI CEO Sam Altman, raising concerns about the company’s safety standards and employment practices toward whistleblowers. The letter, first obtained by The Washington Post, outlines several points, with item 9 being particularly significant: “Will OpenAI commit to making its next foundation model available to U.S. Government agencies for pre-deployment testing, review, analysis, and assessment?”
The letter includes 11 additional points addressing various issues such as dedicating 20% of OpenAI’s computing power to safety research and implementing protocols to prevent AI products from being stolen by malicious actors or foreign adversaries. The lawmakers’ concerns were prompted by whistleblower reports alleging that OpenAI’s safety standards for GPT-4 Omni were compromised to avoid delaying the market release of the product.
Whistleblowers claimed that efforts to raise safety concerns with OpenAI’s management were met with retaliation and allegedly illegal non-disclosure agreements. These concerns were formally filed with the U.S. Securities and Exchange Commission （SEC） in June 2024.
Regulatory scrutiny
The letter from the lawmakers highlights the ongoing regulatory scrutiny faced by OpenAI and the broader artificial intelligence sector. In July, tech giants Microsoft and Apple renounced their memberships on OpenAI’s board due to increased regulatory pressure, despite Microsoft’s $13 billion investment in OpenAI in 2023.
Existential fears persist
Former OpenAI employee William Saunders revealed that he left the company due to concerns that ongoing research might pose an existential threat to humanity. Saunders likened the potential trajectory of OpenAI to the infamous crash of the RMS Titanic in 1912. His concerns were not with the current iteration of OpenAI’s ChatGPT large language model but with future versions and the potential development of advanced superhuman intelligence.
解説
- Government oversight: The letter seeks to establish a framework for government agencies to review and test AI models before deployment, aiming to ensure safety and compliance.
- Whistleblower protection: The lawmakers are addressing concerns about OpenAI’s treatment of whistleblowers, emphasizing the need for transparency and accountability in AI development.
- Regulatory challenges: Increased scrutiny from regulatory bodies reflects the growing importance of ethical and safe AI practices, particularly in light of rapid advancements in AI capabilities.
- Existential risks: The concerns raised by former employees and whistleblowers highlight the potential risks associated with unchecked AI development and the need for robust safety measures.