In a move reflecting its ongoing attempts to balance rapid innovation with security requirements, leading artificial intelligence company OpenAI announced on Monday the launch of the “Safety Fellowship,” a new initiative designed to invite researchers, engineers, and practitioners from outside the company to conduct independent research on the safety of advanced AI systems and their alignment with human values. This ambitious program is scheduled to run from September 2026 through February 2027.
The program provides participants with a wide range of benefits including a monthly stipend, advanced computing resources, guidance and mentorship from company experts and staff, as well as a dedicated workspace at the Constellation research hub in Berkeley, California, with a remote work option available for those who prefer it.
Safety fellowship focus areas and eligibility requirements
Participating researchers in the fellowship will focus on critical areas including safety evaluations, ethics, system robustness, privacy-preserving security methods, autonomous independent oversight, and the study of high-risk misuse areas. The company stated that it seeks to attract applicants from highly diverse backgrounds, including computer science, social sciences, cybersecurity, privacy, and human-computer interaction.
This program is considered a golden opportunity for ambitious researchers seeking to make a real impact on the future of technology, as the company covers all technical and logistical facilities. OpenAI emphasized that research capability and sound technical judgment will be given top priority over traditional academic credentials. At the end of the program, participants are expected to deliver tangible research outputs, such as academic papers, benchmark standards, or open datasets that serve the broader AI community.
Safety challenges and criticisms of commercialization acceleration
This initiative comes at a time when the company faces ongoing questions over whether its pursuit of rapid commercialization for its products has overshadowed its core commitments to safety and security. Earlier this year, the Canadian government urged the company to strengthen its safety protocols after it was revealed that the company failed to notify law enforcement authorities about an account linked to a suspect in a school shooting incident.
Additionally, former company researchers accused management of cherry-picking and publishing only positive safety results that serve its interests, raising concerns about a broader industry trend toward easing strict preventive safeguards, which has drawn widespread scrutiny from regulators and civil society.
A new economic vision for the age of AI
The company’s moves are not limited to technical aspects, but extend to the economic and political landscape. The fellowship is part of a broader political direction. On the same day, the company released a 13-page document titled “Industrial Policy for the Age of Intelligence: Ideas for Human-Centric Governance,” which puts forward proposals for governments to manage economic disruptions caused by increasingly powerful artificial intelligence.
The document calls for establishing a public wealth fund that grants citizens a direct stake in AI-driven economic growth. It also proposes taxing automated labor, supporting the implementation of a four-week-four-day work schedule (editor note: four-day workweek), and expanding social safety nets funded by shifting the tax burden from human labor to robotic capital.
“We are entering a new phase of economic and social regulation that will radically reshape work, knowledge, and production.”
This is how the company described the contours of the next phase in its document, noting that these proposals represent a starting point for open debate rather than final binding recommendations. These proposals reflect a growing recognition that the accelerated development of artificial intelligence will not only affect technical aspects, but will trigger an earthquake in the structure of the labor market and the global economy.
Massive research grants and mixed reactions
To encourage work on these ideas, the company is establishing additional fellowships and offering research grants worth up to $100,000 USD, alongside API programming credits worth up to $1 million USD for projects building on the policy ideas in the document. It also plans to hold a workshop in Washington, D.C. this coming May to gather stakeholders and broaden the discussion.
Despite these moves, critics have questioned the true motives behind the framework put forward by the company. Press reports, such as those published by Fortune magazine, indicated that some observers view these policy proposals as a form of “regulatory nihilism”—an attempt to shape regulations that primarily serve the company’s interests while pretending to act for the public good. Analysts point out that the overlap between the worlds of finance and business on one hand, and government legislation on the other, makes it difficult to accept these proposals without rigorous scrutiny of their hidden motives.
For its part, The Wall Street Journal noted that the timing of these proposals coincides with Congress preparing to debate crucial legislation concerning artificial intelligence, and at a time when the Trump administration is seeking support from the tech sector ahead of the elections. In the same context, the Gizmodo platform described the document as vague and lacking details, noting that the company has echoed such ideas for years without offering practical solutions or clear steps on what any entity should do to address these complex challenges.
Frequently Asked Questions
What is the safety fellowship recently launched by the company?
It is a research program that brings together experts and researchers from outside the company to study AI safety, ethics, and privacy. The program will run from September 2026 to February 2027.
What are the most prominent economic proposals in the company’s policy document?
The company proposed creating a public wealth fund, taxing automated labor, reducing weekly workdays to four days, and expanding social safety nets to protect individuals from labor market disruptions.
Why is the company facing increasing criticism regarding safety policies?
Some researchers and critics accuse the company of accelerating the commercial launch of its products at the expense of safety standards, alongside accusations of cherry-positive data when publishing safety reports and ignoring critical incidents such as the one that occurred in Canada.
What is the value of the grants the company will provide to support policy research?
The company plans to provide research grants of up to $100,000 USD, along with API credits of up to $1 million USD for projects building upon the ideas of the policy document.
