Back to list
Mira Murati Testifies Under Oath Regarding Sam Altman’s Alleged Misrepresentations on AI Safety Standards
Industry NewsOpenAISam AltmanMira Murati

Mira Murati Testifies Under Oath Regarding Sam Altman’s Alleged Misrepresentations on AI Safety Standards

Former OpenAI CTO Mira Murati has provided sworn testimony in the ongoing Musk v. Altman trial, alleging that CEO Sam Altman misled her regarding the safety protocols of a new AI model. In a video deposition, Murati stated that Altman falsely claimed OpenAI's legal department had determined a new model met safety standards when it had not. This testimony highlights significant internal friction and a breakdown of trust at the highest levels of OpenAI leadership. The revelation comes at a critical time as the industry faces increasing scrutiny over AI safety governance and executive transparency. Murati’s statements under oath suggest that internal verification processes for AI safety may have been misrepresented within the organization's executive tier.

The Verge

Key Takeaways

  • Sworn Testimony: Former OpenAI CTO Mira Murati testified under oath that she could not trust the words of CEO Sam Altman.
  • Allegations of Deception: Murati alleged that Altman lied specifically about the safety standards and legal clearance of a new AI model.
  • Legal Department Dispute: The testimony claims Altman falsely stated that the legal department had approved the safety of a model.
  • Trial Context: These statements were revealed via a video deposition during the Musk v. Altman trial proceedings.

In-Depth Analysis

The Breakdown of Executive Trust and Transparency

The testimony provided by Mira Murati represents a significant moment in the public discourse surrounding OpenAI’s internal governance. By stating under oath that she could not trust Sam Altman’s words, Murati points to a fundamental rift in the relationship between the Chief Technology Officer and the Chief Executive Officer. This lack of trust is not merely personal but is tied directly to the technical and ethical oversight of the company’s products. When the individual responsible for the technical roadmap of an AI organization feels misled by the chief executive regarding safety, it suggests a systemic issue in how safety information is communicated and verified internally.

Murati’s deposition highlights a specific instance of alleged misrepresentation. According to her testimony, Altman claimed that OpenAI's legal department had conducted an assessment and determined that a new AI model was in compliance with safety standards. Murati asserts that this statement was false. This specific allegation is critical because it involves the intersection of technical safety, legal compliance, and executive reporting. If safety determinations are being misrepresented to top-level technical staff, it raises questions about the integrity of the entire safety framework within the organization.

Safety Standards and Internal Verification Processes

The core of the dispute mentioned in the trial involves the "safety standards for a new AI model." In the high-stakes environment of artificial intelligence development, safety standards are the primary safeguard against unforeseen risks. Murati’s testimony suggests that these standards may have been bypassed or that the status of their verification was inaccurately reported to ensure the progression of the model.

The mention of the "legal department" is particularly noteworthy. In many technology firms, the legal department acts as a final checkpoint for compliance and risk management. By allegedly claiming that the legal department had already cleared the model, Altman would have effectively neutralized one of the primary internal hurdles for the model's deployment or further development. Murati’s challenge to this claim under oath indicates that the internal checks and balances designed to ensure AI safety may have been compromised by executive narrative-building rather than factual technical or legal consensus.

Industry Impact

The implications of Murati’s testimony extend far beyond the walls of OpenAI. As one of the most influential figures in the AI industry, her public declaration of a lack of trust in Sam Altman’s transparency regarding safety could lead to several shifts in the broader AI landscape:

  1. Increased Regulatory Scrutiny: Regulators may view this testimony as evidence that self-regulation within AI companies is insufficient. If a CTO cannot rely on the safety claims made by a CEO, government bodies may feel compelled to implement more rigorous, independent third-party auditing of AI safety standards.
  2. Focus on Governance Structures: The AI industry may see a push for more robust governance structures where safety and legal departments report to independent boards rather than directly to the CEO, preventing the potential for misrepresentation of safety data.
  3. Impact on Corporate Culture: This high-profile legal battle and the resulting testimony may influence how other AI startups handle internal transparency. It serves as a cautionary tale regarding the long-term legal and reputational risks of internal communication breakdowns concerning safety protocols.

Frequently Asked Questions

Question: What did Mira Murati specifically allege about Sam Altman in her testimony?

Answer: Mira Murati testified that Sam Altman lied to her regarding the safety standards of a new AI model, specifically claiming that the legal department had determined the model met safety requirements when, according to Murati, that was not the case.

Question: In what legal proceeding did this testimony come to light?

Answer: The testimony was part of a video deposition shown during the ongoing Musk v. Altman trial on Wednesday.

Question: Why is the mention of the legal department significant in this testimony?

Answer: It is significant because it suggests that the CEO may have used the perceived authority of the legal department to bypass or misrepresent safety concerns to the CTO, indicating a potential failure in the company's internal verification and compliance processes.

Related News

How Google Duplex Evolved to Call Businesses on Your Behalf in Gemini
Industry News

How Google Duplex Evolved to Call Businesses on Your Behalf in Gemini

A Google Research Blog entry titled “How Google Duplex evolved to also call businesses for you in Gemini” describes its subject as the evolution of Google Duplex to support calls to businesses on a user’s behalf in Gemini. The entry is dated November 20, 2026, and the supplied article text consists only of the label “Generative AI.” The headline identifies the capability and connects it to Duplex, but the available text does not explain how the system evolved, how users request a call, or where the feature is available. The headline establishes the topic, but the supplied material does not provide a technical account or rollout details.

How AI and Lean Are Helping Automate Formal Verification to Fight Software Slop
Industry News

How AI and Lean Are Helping Automate Formal Verification to Fight Software Slop

In Latent Space, Richard MacManus examines formal verification as an emerging way to fight what the headline calls “slop.” The central claim is that proving software correct with mathematics has traditionally required specialist work, and that AI and Lean are now helping automate that process. The distinction is important: the reported change concerns how the work of proving correctness is done, not what counts as a proof. The supplied excerpt presents automation as assistance, without saying that it replaces specialists or completes proofs independently. It also does not specify which parts of the process AI and Lean help automate, how much work they save, or which software has been verified this way.

AI and Human Claim Solutions to the 167-Year-Old Riemann Hypothesis
Industry News

AI and Human Claim Solutions to the 167-Year-Old Riemann Hypothesis

TechCrunch AI reports that AI claimed a solution to the Riemann hypothesis, a mathematical problem the headline describes as 167 years old. A human also put forward a competing solution. The report, written by Tim Fernholz and published on November 20, 2026, describes mathematicians as astonished by the AI-driven result. The supplied excerpt does not explain either proposed solution, identify the AI system or the human claimant, or say whether either claim has been verified. The distinction between a claimed solution and a confirmed one is central to reading the report: the excerpt establishes that two claims are being reported, but it does not establish that either has resolved the hypothesis.