news

OpenAI, a pioneer in artificial intelligence, seems to have fallen into some trouble recently. The company's successive core senior personnel changes have attracted widespread attention in the industry.

Especially in recent months, there has been a new wave of resignations. Currently, only 2 of the 11 people remain in the company.

President Greg Brockman is one of them., saying "This was co-created 9 years agoOpenAIThe first time I relaxed after that.”

The whereabouts of former research director Ilya Sutskever has sparked further speculation.Safe Superintelligence Inc., which claims to be focused on one goal: achieving safe superintelligence through revolutionary breakthroughs, coincides with the original intention of the OpenAI nonprofit organization.

Since its establishment, OpenAI's non-profit identity has always attracted much attention. But now the tension between its non-profit mission and commercial transformation has become increasingly apparent, causing core team members to leave one after another. Some analysts also pointed out that the reason for the loss of personnel may also be related to Musk's lawsuit this week.

The shift in nonprofit mission: drifting away from the company’s original purpose

At the end of 2015, OpenAI announced its founding to the world. The declaration was only 600 words long and began with:

OpenAI is a nonprofit artificial intelligence research company. Our goal is to advance digital intelligence in the way that is most likely to benefit all of humanity, unconstrained by the need to generate financial returns. Because our research is free of financial obligations, we can better focus on having a positive impact on humanity.

We believe that AI should be an extension of the individual human will and, in the spirit of freedom, should be distributed as widely and evenly as possible. The outcome of this adventure is uncertain and the work is hard, but we believe the goals and structure are the right ones. We hope this is the most important for the best in the field.

The two authors of this article are Greg Brockman, who is on leave, and Ilya Sutskever, who has resigned.

Today's OpenAI seems to have drifted further and further from this original declaration, just like Sam Altman and other founding members.

In 2019, OpenAI announced the establishment of OpenAI LP, a capped-profit company, to attract external investment. This structure allows investors to receive certain returns while limiting the company's profitability, thereby maintaining its original non-profit nature.

Perhaps it was this shift that caused tension within the company, with many original team members leaving due to dissatisfaction with OpenAI's increasingly commercial direction.As Sam Altman gained more power in OpenAI, he began to have the ability to drive OpenAI forward with his own ideas.

Of course, most companies have to evolve as they go. But most companies don’t start out with such a clear idealistic flag — and the results are less than ideal. Of course, OpenAI is still a nonprofit organization in some ways, as it’s not yet profitable.

Some analysis also pointed out that the reason for the loss of personnel may also be related torelated.This week, Musk filed a lawsuit against Altman in federal court, alleging that he was misled into co-founding OpenAI.

Musk claims he invested in OpenAI based on the promise that it would operate as a nonprofit, but later discovered that Altman and partners such as Microsoft had created a series of opaque for-profit subsidiaries and engaged in serious self-dealing. The new lawsuit is similar to another lawsuit Musk filed against OpenAI and Altman in California earlier this year, which he eventually withdrew.

Financing pressure, GPT-5 difficulties, pressure from competitors... What is the future of OpenAI?

Like other AI startups that are burning money, OpenAI is currently facing huge financial pressure. Many smaller AI companies have chosen to be acquired by large companies, but Sam Altman has been raising large amounts of funds. Despite Microsoft's $1 billion backing, the relationship between the two companies seems to be unstable recently, the "honeymoon period" has broken down, and the situation is becoming more and more complicated.

Following the exitLast week, Microsoft officially launched OpenAI。

Some analysts believe that OpenAI may need to find new, large investors, such as Apple, Saudi Arabia or SoftBank.

At the same time, GPT-5 seems to have encountered a "difficult birth". There are reports that the development of GPT-5 is much slower than expected. Although everyone hopes to see results before the end of the year,But at this fall's developer conference, officials have made it clear that no new models will be released.

Techcrunch believes that in recent months, OpenAI has taken more incremental steps, including focusing on optimizing existing models. GPT-4o However, as high-quality training data becomes more difficult to obtain, OpenAI's technical lead in the generative AI race may be weakened.

Pressure also comes from strong competitor Meta.Meta's Zuckerberg publicly stated that the computing resources required to train the next-generation model Llama 4 are 10 times that of the previous generation. This reveals the huge cost of training large models and also hints at the financial pressure OpenAI may face - because Meta is a for-profit company, while OpenAI is still a non-profit.

Developing advanced AI models requires huge resources and time, and the competition between open source and closed source models is becoming increasingly fierce. More and more people, including Zuckerberg, believe that open source AI models (such as Meta's Llama) are better than closed source models (such as OpenAI'sGPT) has more advantages.

Now, OpenAI is at a critical crossroads. Whether the company's core competitiveness can still maintain its advantage has become the focus of the industry. In addition, the ultimate goal of the field of artificial intelligence - general artificial intelligence (AGI) is still the company's pursuit, or whether to turn to more commercial product development has become an important choice facing OpenAI.

Appendix: OpenAI leadership changes

Greg Brockman: On leave starting August 2024

Brockman was a core member of OpenAI's founding team. At the persuasion of Altman and Musk, he resigned from his job as CTO of the fintech company Stripe to take the same position at OpenAI. He has been a key ally of Altman from the beginning. When the board turned against Altman in a coup in November, Brockman was also removed as a director. Five days later, when the board withdrew, the two returned to their positions together. On Monday, he announced that he would take a leave of absence for the rest of the year. "For the first time since co-founding OpenAI 9 years ago, I am relaxing," he wrote on X. "For the past 9 years, I have poured my life into OpenAI, including my entire marriage. Our work is important to me, but so is life."

John Schulman: Joining Anthropic in August 2024

Schulman, who works at the development companyChatGPTchatbot, announced on Monday that he is leaving OpenAI. He was responsible for fine-tuning the company's AI models and ensuring that their behavior is consistent with human values ​​​​— a process known as "alignment." He will take a similar role at rival startup Anthropic, which was itself founded by former OpenAI researchers in 2021. "This choice is driven by my desire to focus deeply on the AI ​​alignment problem while starting a new chapter in my career and re-engaging in practical technical work with people who are deeply engaged in the topics that interest me most," Schulman said in a note to colleagues on Monday.

Ilya Sutskever: Leaves in May 2024 to found Security Super Intelligence

Sutskever left his position as chief scientist at OpenAI six months after voting with the company's board to oust Altman. Sutskever, one of the most prominent researchers in the field, changed his stance a few days later, supporting the return of the CEO. However, he has largely disappeared from public view since that failed change, and left in May to start a company.

Andrej Karpathy: Founded Eureka Labs in February 2024

Karpathy, a research scientist who was advised by "AI godmother" Fei-Fei Li at Stanford University, first left OpenAI in 2017 to join Tesla as a senior director. He returned to OpenAI in 2023 and left again a year later to found Eureka Labs, which is training AI teaching assistants.

Durk Kingma: Left in June 2018 to join Google Brain

Kingma, who worked on developing algorithms for generative AI models, left OpenAI for Google in the summer of 2018. He continues to lead research on large language models and image models in the Google Brain team, which last year partnered withDeepMindmerge.

Elon Musk: Resigned from the board in 2018

After providing a large portion of OpenAI’s early funding, Musk left the company in 2018 amid a conflict with Altman over research direction. The billionaire founded a competing company called xAI last year and claims he can overtake OpenAI’s lead.

The head of Tesla, SpaceX and X has also launched multiple lawsuits against Altman and OpenAI, and said this week that he was induced to invest in the AI ​​company because of its "false humanitarian mission."

Pamela Vagata: Joined STRIPE in 2016

Vagata, who is listed as a founding member of OpenAI in the company’s launch announcement but does not mention the startup on her LinkedIn profile, joined Stripe in 2016 as technical lead for the fintech company’s AI team and founded early-stage venture capital firm Pebblebed in 2021.

Vicki Cheung: Joined Lyft in 2017

Before becoming OpenAI’s first engineer, Cheung worked at language learning app Duolingo, which she left in 2017 to join ride-hailing startup Lyft. In 2020, she founded machine learning startup Gantry with former OpenAI researcher Josh Tobin.

Trevor Blackwell: Left in 2017

Blackwell was a partner at Y Combinator, the San Francisco startup accelerator that Altman ran before founding OpenAI. He helped start the artificial intelligence company and left in 2017. A robotics enthusiast, he now lives in Gloucestershire, England.

Current personnel:

Sam Altman

Altman continued to serve as OpenAI’s CEO after surviving a board coup in November, during which directors accused him of not being “consistently candid” with them. He was reinstated after a campaign by OpenAI employees and investors, including Microsoft.

The departures of other senior figures have made the 39-year-old the most prominent figure at the company by far, while a board reshuffle following a failed bid to oust him has further consolidated his power.

Wojciech Zaremba

Zaremba, a Polish computer scientist, remains at OpenAI as a researcher. He called on the board to resign after they turned against Altman and urged his CEO and Musk to abandon their "unnecessary fight." "In the midst of quarrels, it is better to invest your creativity in building the future of your dreams. May you (both) be happy and find peace," he wrote in the March X post, signing off with a heart.