The Midas Project, an artificial intelligence watchdog, has filed a formal complaint with the U.S. Internal Revenue Service against OpenAI, the creator of ChatGPT, alleging potential violations of federal tax laws that could jeopardize the company’s nonprofit status.
In its complaint, the watchdog accuses OpenAI of dismantling internal safeguards, fostering significant conflicts of interest, and potentially misusing charitable resources. It also highlights that OpenAI CEO Sam Altman could personally benefit if the organization leans further into its for-profit ventures.
According to the Midas Project, OpenAI’s governance structure poses serious ethical and legal concerns, particularly due to Altman’s dual role as CEO of its for-profit subsidiary and board member of the nonprofit entity. The complaint argues this setup creates a conflict in which Altman stands to gain financially at the nonprofit’s expense.
“Sam Altman may receive equity in OpenAI’s for-profit arm, and his private investments in companies partnered with OpenAI raise additional conflicts of interest potentially worth hundreds of millions,” the group stated.
Founded in early 2024, the Midas Project describes itself as a nonprofit organization focused on holding leading AI companies accountable and ensuring that artificial intelligence serves the public good—not just the interests of the corporations developing it.
Conflicts of interest alleged
According to the New York Post, Sam Altman is expected to receive an equity stake in a restructured OpenAI entity—an interest that could be worth billions, given the company’s estimated $300 billion valuation.
The Midas Project’s complaint also highlights financial conflicts involving several OpenAI board members. Among them are Chairman Bret Taylor, co-founder of Sierra AI, a company that resells OpenAI’s models; Adam D’Angelo, whose platform Quora is a customer of OpenAI; and Adebayo Ogunlesi, whose firm, Global Infrastructure Partners, owns data centers positioned to profit from the growing demand for AI infrastructure.
OpenAI was originally founded as a nonprofit with the mission to ensure that artificial general intelligence (AGI) would serve the interests of humanity—not private shareholders, the watchdog emphasized. The Midas Project concluded its complaint by arguing that OpenAI’s current trajectory risks betraying that founding mission.
“As they race toward powerful AI systems, any weakening of these protections could put the public interest at risk. The IRS must investigate to preserve OpenAI’s duty to humanity.”
Profit or nonprofit?
OpenAI was established in 2015 as a nonprofit organization by Elon Musk, Sam Altman, and others with the goal of developing artificial intelligence for the benefit of humanity.
In November, the company reportedly entered preliminary discussions with U.S. regulators about transitioning to a for-profit structure.
However, those plans were later put on hold. In May, OpenAI announced that it had abandoned the restructuring efforts for the time being and reaffirmed its commitment to its nonprofit mission, stating that no final decisions had been made.
The Elon Musk connection
In 2017, Elon Musk initially pushed for OpenAI to transition into a for-profit entity. However, in both March and August 2024, he filed lawsuits against the company, accusing it of breaching its original nonprofit agreement and straying from its founding mission in pursuit of profit.
Earlier that year, in February, a group of investors led by Musk submitted a $97.4 billion offer to acquire OpenAI. The bid was ultimately rejected by CEO Sam Altman and the board.
