OpenAI’s attempt to keep Stuart Russell out of its ongoing court battle with Elon Musk has opened a second argument beyond the lawsuit itself: how seriously the company treats the AI safety warnings it has used in public for years.
According to The Midas Project, OpenAI’s latest court position creates a sharp contradiction. The company is challenging Russell as a witness while distancing itself from concerns that closely resemble warnings OpenAI and Sam Altman have previously endorsed.
The Court Fight Over Stuart Russell
OpenAI filed a motion to exclude testimony from AI professor Stuart Russell. In that filing, the company described him as a "prominent AI doomer" and said he has "made a career giving public lectures warning that AI might kill off humanity." OpenAI’s lawyers also characterized his views as "dystopian," "speculative," and "alarmist."
Russell’s position, as described in the source article, is that advanced AI systems could pose an existential threat to humanity if they are not properly controlled. That view is central to the dispute raised by The Midas Project because OpenAI has not always treated those concerns as fringe or irrelevant.
The Midas Project argues that OpenAI is using courtroom language that conflicts with the company’s own public history. The issue is not only whether Russell should testify. It is whether OpenAI can dismiss AI extinction risk as speculative in court after building parts of its public identity around similar concerns.
Why The Midas Project Sees A Double Standard
The Midas Project says OpenAI has long relied on serious AI risk scenarios to support its brand and policy agenda. The source article points to warnings about "scary" mass job losses, social manipulation, and doomsday scenarios as examples of the kind of rhetoric OpenAI has previously used.
That history matters because the company is now pushing back against testimony from a researcher associated with existential risk concerns. The Midas Project’s criticism is that OpenAI appears to accept these warnings when they support its own narrative, but frames them as disqualifying when they appear in an opposing legal context.
The contrast becomes more pointed because Sam Altman co-signed a declaration with Russell in 2023 stating that the risk of extinction from AI should be a global priority. The source article also notes that Altman said back in 2015 that AI would "probably, most likely, sort of lead to the end of the world."
Those details make the dispute more than a narrow fight over expert testimony. They put OpenAI’s public AI safety messaging beside its legal strategy and ask whether the same concerns are being treated consistently.
The AGI Race Argument
OpenAI’s lawyers also argue that worries about an AI race at the expense of safety are not relevant to the case. The Midas Project counters that OpenAI’s own charter describes a similar concern: that the late stages of AGI development could become a race without enough safety precautions.
In plain terms, the disagreement is about relevance. OpenAI’s legal team is trying to limit what the court considers. The Midas Project argues that the concern OpenAI wants to push aside is already part of OpenAI’s stated worldview.
The source article says OpenAI may already be showing signs of that pattern. It says safety testing for GPT-4o was reportedly rushed under time pressure, and that newer models have been affected too. It also says the company cut jobs on its safety teams and that several high-profile researchers left.
These points are presented as part of The Midas Project’s broader claim: if OpenAI has warned that competitive pressure could weaken AI safety, then arguments about that pressure should not be dismissed as unrelated when they arise in court.
A Wider Conflict With The Midas Project
The dispute between OpenAI and The Midas Project did not begin with the Stuart Russell motion. The source article says OpenAI previously hit The Midas Project with a broad subpoena as part of the Musk dispute, even though the organization was not a party to the lawsuit.
According to founder Tyler Johnston, OpenAI demanded all communications with journalists, lawmakers, former OpenAI employees, and other civil society groups. The Midas Project viewed that demand as significant because it had publicly criticized OpenAI’s restructuring.
Johnston suspected the subpoena was timed around decisions by California and Delaware authorities on OpenAI’s conversion to a for-profit company with a valuation of up to $500 billion. That claim adds another layer to the conflict: The Midas Project is not only objecting to OpenAI’s arguments about Russell, but also to how the company has treated critics during the broader dispute.
What The Fight Reveals About AI Safety Messaging
The core tension is simple. OpenAI has publicly treated advanced AI risk as a major issue. In court, according to The Midas Project, it is trying to narrow or discredit similar concerns when they come from Stuart Russell.
That does not settle the legal question of whether Russell’s testimony should be admitted. But it does raise a reputational question for OpenAI: can a company warn about AI extinction risk, cite competitive safety dangers in its own charter, and then describe a prominent safety expert’s warnings as too speculative when they become inconvenient?
For readers following AI safety, the case shows how public principles can be tested in legal settings. The same ideas that sound urgent in a declaration, a charter, or a public warning may be treated very differently when they affect litigation strategy.
The Midas Project’s criticism is that OpenAI is making opportunistic arguments. Based on the facts described in the source article, the organization sees a mismatch between OpenAI’s past safety claims and its present courtroom posture. That mismatch is now part of the public debate around OpenAI, Stuart Russell, and the company’s evolving approach to AI risk.