OpenAI has changed its mission statement 6 times in 9 years, most recently about AI that "safely benefits humanity"

OpenAI has made its sixth mission statement revision in nine years, with the most recent change removing the critical qualifier 'safely' from its pledge to benefit humanity with artificial general intelligence. This alteration was documented in the company's final IRS Form 990 filing as a tax-exempt nonprofit, covering fiscal year 2024 and released in November 2025. The shift comes in the wake of OpenAI's major corporate restructuring, which transferred nearly three-quarters of the nonprofit's control to a for-profit entity owned by private investors and employees, fundamentally changing its governance and incentive structure.

The removal of 'safely' is not merely semantic; it reflects a concrete change in the company's legal and financial priorities. With investors now holding board seats and directly receiving a share of OpenAI's profits, there is growing concern that safety research and long-term risk mitigation could be deemphasized in favor of accelerating product development and maximizing financial returns. This evolution from a pure research lab to a commercially-driven entity marks a pivotal moment, raising questions about how the company will balance its original founding principles with the pressures of its new for-profit reality.