OpenAI Shifts Priorities: Disbanding the Safety Team Raises Concerns Ahead of IPO

At the end of July 2026, the tech giant OpenAI made a decision that sparked many questions and some alarm in expert circles: the company disbanded its dedicated “preparedness” team. This unit played a key role in identifying and minimizing potential risks associated with the development and deployment of advanced artificial intelligence models. The decision, based on information from internal sources, signals significant changes in the company’s strategy, especially given its ambitious plans to go public.

The “preparedness” team was one of the last elements of OpenAI’s old structure, which now appears to be undergoing a deep transformation. Before that, in 2024, Jan Leike, who co-led the superalignment team, announced his resignation. His decision was motivated by his statement that safety, in his view, had taken a back seat to marketing campaigns and “nice covers.” This trend is concerning, as the risks associated with the rapid development of AI require constant vigilance and a proactive approach.

A shift in focus: from safety to commercial success?

According to information circulating among OpenAI employees, the decision to disband the “preparedness” team was made at the direct instruction of CEO Sam Altman. His strategic direction calls for cutting back so-called “side quests” and maximizing efforts on developing ChatGPT and aggressively competing with other leading AI developers, especially Anthropic, for enterprise customers. This points to a clear shift from theoretical safety concerns to the practical realities of commercialization and market-share expansion.

Although the “preparedness” team itself was eliminated, its key specialists did not lose their jobs. Senior staff from the unit were moved into responsible roles within existing teams. They were assigned to oversee specific safety areas such as biosecurity and cybersecurity. At first glance, this resource reallocation may seem like a logical step to integrate expertise. However, given previous statements and the broader trend, questions remain about whether these critical areas will receive enough attention under the new organizational structure.

These changes are taking place against the backdrop of a noticeable wave of departures among top OpenAI executives. Since 2026, the company has lost a number of key managers, which naturally raises questions about stability and long-term strategy. Among those who have left the company are:

• Denise Dresser, the chief business officer responsible for attracting enterprise clients. Her departure after less than a year on the job is surprising, especially given how important this role is in the context of IPO preparation.
• Chloe Bacalar, head of ethics, resigned in July 2026, also after less than a year. This may indicate some differences in views on the ethical dimensions of AI development.
• Brad Lightcap, former chief operating officer (COO), who after eight years at OpenAI announced his resignation in August 2026. His transfer to a new, “loosely defined” role before leaving, as in other cases, raises questions.
• Joshua Achiam, chief futurist and former head of mission alignment, left in July 2026 after nearly nine years. His departure, like the others, was tied to a transfer to another role.
• Fidji Simo, once described as OpenAI’s “No. 2,” stepped back from active duties in July 2026 and moved into an advisory role.
• Kevin Weil, former chief product officer who led OpenAI for Science, left in April 2026.
• Srinivas Narayanan, the chief technology officer (CTO), who had spent the last three years at OpenAI, also resigned in April 2026.

It is worth noting that many of these top executives were moved into new or vaguely defined roles before their departures. This practice often suggests an attempt by the company to maintain some appearance of stability, or an inability to find them a suitable place in the new structure.

IPO on the horizon: is OpenAI ready for the challenge?

These personnel changes are unfolding amid OpenAI’s active preparations to go public, which, according to the Financial Times, could give the company a valuation of around $1 trillion. Despite the ambitious plans, the internal chaos that is visible is raising concerns among experts. The IPO was originally planned for this year, but according to informed sources, it will likely be pushed back to next year.

In the artificial intelligence market, OpenAI faces fierce competition. Its main rival remains Anthropic, which is showing impressive growth rates. While OpenAI’s revenue doubled over the past year, Anthropic’s figures increased fivefold, underscoring its dynamic expansion and the effectiveness of its strategy.

On June 9, 2026, the developer of ChatGPT officially filed a confidential application for an initial public offering (IPO) with the U.S. Securities and Exchange Commission. This was an important step toward becoming a public company. Among other giants also preparing to go public, it is worth mentioning Anthropic and SpaceX, which have already taken similar steps by filing confidential applications.

The AI market is experiencing rapid growth, and the public listing of major players such as OpenAI could significantly influence global investment flows. However, questions of safety, ethics, and internal stability remain key challenges, and OpenAI’s long-term success depends on how well it addresses them. The disbanding of the team responsible for assessing and minimizing risks during the company’s active IPO preparations will undoubtedly draw close attention from investors and analysts.