Congressional Democrats have sent a letter to OpenAI CEO Sam Altman, requesting detailed information on the company’s safety standards and its treatment of whistleblowers.

Key Requests from Lawmakers

The letter, initially reported by The Washington Post, features a crucial question in item 9: “Will OpenAI commit to making its next foundation model available to U.S. Government agencies for pre-deployment testing, review, analysis, and assessment?”

Additionally, the letter outlines 11 other points for OpenAI to address, including:

• A commitment to allocate 20% of its computing resources to AI safety research.
• Protocols to prevent malicious actors or foreign adversaries from stealing AI products from OpenAI.

First page of the letter from Senate Democrats. Source: The Washington Post

Increased Regulatory Scrutiny

OpenAI and the broader AI industry have long faced regulatory scrutiny. The latest letter from Democratic lawmakers was prompted by whistleblower reports alleging that OpenAI compromised on safety standards for GPT-4 Omni to avoid delaying its market release.

Whistleblowers within OpenAI claimed that their safety concerns were met with retaliation and that they were forced to sign allegedly illegal non-disclosure agreements. These concerns led to a complaint filed with the U.S. Securities and Exchange Commission in June 2024.

Shortly after these developments, tech giants Microsoft and Apple withdrew their memberships from OpenAI’s board due to the increased regulatory pressures. This decision came despite Microsoft’s $13-billion investment in OpenAI in 2023.

Persistent Existential Fears

Former OpenAI employee William Saunders revealed that he resigned from the company due to fears that its ongoing research could pose an existential threat to humanity. Saunders compared the potential risks of OpenAI’s trajectory to the infamous crash of the RMS Titanic in 1912.

While Saunders was not immediately concerned about the current iteration of ChatGPT, he expressed worries about future versions and the potential development of advanced superhuman intelligence.

He emphasized that employees in the AI sector should have the right to warn the public about the potentially dangerous capabilities arising from rapid advancements in synthetic intelligence.

Source: Cointelegraph