According to a Reuters report, OpenAI, the creator of ChatGPT, plans to restructure its core business into a commercial corporation. After this process is over, the for-profit entity will no longer be controlled by the nonprofit’s board of directors. OpenAI CEO Sam Altman will get a stake in a commercial company for the first time.

OpenAI is one of the key players in the new summer of artificial intelligence. In November 2022, the company launched the ChatGPT service, in which a large language model answers user questions in text chat mode. The huge popularity of this and other products has brought the relatively small OpenAI a lot of money. According to the New York Times, in the current round of investments, the company raised $6.5 billion for a total valuation of about $150 billion.

However, OpenAI is far from an ordinary startup. OpenAI was founded in December 2015 by Elon Musk and Sam Altman as a non-profit organization. At the time of its inception, the non-profit’s website looked like a simple web page with a minimal design, with a stated goal: to create universal artificial intelligence that is safe and useful for humanity. This press release hung on the openai.com website for the first few months.

Since its inception, OpenAI has received investments. Do not overestimate their size. In total, in the first three years, $130 million was poured into the project, which had to be spent on the salaries of highly paid machine learning specialists — in 2017, there were 45 employees.

A part of OpenAI’s resources was taken “in kind”, but even in this case, the size of the gifts was not impressive. For example, in 2016, the head of Nvidia, Jensen Huang, donated the DGX-1 computer for neural network models to OpenAI. The only computer was taken personally by Elon Musk. Amazon and Microsoft have provided computing power in their clouds for free. Someone bought control over the organization: the Open Philanthropy fund gave ₽30 million for a seat on the board of OpenAI for 3 years.

In 2018, as a result of disagreements with Sam Altman, Elon Musk left the board of directors of OpenAI, whose total fortune that year was estimated at $22.5 billion. The billionaire himself said that he gave OpenAI $50 million or even close to $100 million, although the TechCrunch analysis comes to the conclusion that Musk brought in only $15 million.

In any case, already in 2019, OpenAI announced that it was forming a commercial subsidiary, OpenAI LP. The description of the complexities of the organization occupies a separate page on the OpenAI website. The structure is built so that all control levers over the commercial part remain with the OpenAI Nonprofit organization and its board of directors. The concept of profit limitation is introduced [capped profit] for investors and employees: all money over the limit (negotiated individually for partners) is returned to the Nonprofit.

These were not empty words, OpenAI really maintained a non-commercial structure. In 2019, Microsoft invested in the company, and at the beginning of 2023, the size of this cooperation grew to multibillion. In this, Microsoft formally has little control over OpenAI. Microsoft is often mistakenly credited with a 49% stake in OpenAI, although the tech giant is entitled to 49% of the revenue of OpenAI’s commercial arm.

In November 2023, there was an attempt to remove Altman from the company. Sam was fired from OpenAI, but returned within 5 days. After that, many old board members were gradually replaced. Ilya Sutzkever and Helen Toner left the company. OpenAI co-founder Greg Brockman also lost his board seat and remained with OpenAI as president, but announced in August 2024 that he was taking a leave of absence until the end of the year.

At the moment, the board of directors of OpenAI includes 8 people:

1. Bret Taylor (chairman);

2. Sam Altman (Executive Director);

3. Sue Desmond-Hellman;

4. Nakasone flooring;

5. Nicole Seligan;

6. Fiji Simo;

7. Larry Summers;

8. Zico Colter.

Also, OpenAI is simply losing key employees. In August, co-founder and one of the company’s key researchers, John Shulman, announced his move to Anthropic. A few hours ago, the technical director of OpenAI Miru Murati told about her departure.

Today, citing its own unnamed sources, Reuters reported that OpenAI plans to restructure its core business. The nonprofit will survive, but will have a minority stake in the for-profit company, sources said.

Altman himself will receive shares in the company, according to insiders. The size of his share is not reported. It is known that the current estimate of the value of all OpenAI is 150 billion dollars.

Interestingly, this contradicts Altman’s previous statements. Microblogger Ashatosh Srivastava cites a fragment of the 2023 hearings in the US Senate on the topic of artificial intelligence regulation. Sam Altman took part in the hearing at the time.

Senator John Neely Kennedy asked Altman if he made a lot of money. Sam replied that he earns little (enough for health insurance) and has no stake in OpenAI. To the senator’s surprised exclamation, Altman added that he was in the company because he liked it.

Altman’s net worth without OpenAI is estimated at $1 billion or $2 billion. The entrepreneur founded and sold the company Loopt, headed startup incubator Y Combinator for several years, and now owns stakes in Stripe, Reddit, and nuclear fusion firm Helion.

OpenAI employees receive a Profit Participation Unit bonus, similar to shares, in addition to their basic salary. In a comment for Bloomberg, a representative of OpenAI said that Altman does not have such a PPU. Also, Sam did not invest in the OpenAI Startup Fund, so although he formally owns 75% of it, he will not receive any part of the fund’s income.

Lawyers and shareholders are still working out the details of the restructuring behind closed doors, Reuters said, citing its sources, and a specific time frame has not yet been set.

Reuters points out that OpenAI is trying to get rid of the profit cap for investors. The publication believes that avoiding the control of a non-profit board of directors will make the company look more like a typical startup, which will appeal to investors. However, Reuters notes that the AI ​​security community may have concerns about whether OpenAI will be able to control itself in the race for universal artificial intelligence.

Reuters recalls that OpenAI disbanded the Superalignment team this year to investigate the long-term risks of artificial intelligence. The Superalignment project was started by Ilya Sutzkever and Jan Leicke. If Sutzkever refrained from negative comments about his departure from OpenAI, Leike after moving to Anthropic sharply criticized the culture of his former employer. As Ian stated, in recent years OpenAI has taken a backseat to security issues in favor of attractive products.