Artificial Intelligence: The Mystery of Sam Altman’s Expulsion – Business

Last week Sam Altman announced the future. On a stage in San Francisco, the head of the company OpenAI presented the next step in his master plan: a platform on which artificial intelligence (AI) for every situation in life. From the graphics program that magically converts text into images, to the AI ​​coach for preparing for salary negotiations. They are based on the same technology as Chat-GPT, the world-famous chatbot from OpenAI. The company is building a world full of virtual assistants that serve people. And he, Sam Altman, pushed the coexistence of man and machine forward like no other, it seemed.

Now others will do that. OpenAI’s board of directors kicked Altman out. The company is thus getting rid of one of the best-known faces of the AI ​​boom, which has been affecting large parts of society since November 30th last year. At that time, OpenAI had activated Chat-GPT.

Internet companies are emulating OpenAI

Since then, Google, Meta and now Amazon and Apple have been trying to catch up with their so-called language models, which can represent linguistic connections almost as well as humans can. After a billion-dollar investment from Microsoft, OpenAI is focusing on expanding its technical lead over the competition with the help of Microsoft’s cloud infrastructure. The AI ​​race is running at an enormous expense of money and unprecedented computing capacity. Altman traveled around the world to convince politicians, entrepreneurs and journalists that AI was as revolutionary as the smartphone.

Virtually no one in the AI ​​industry expected him to be fired. The opinion reads brutally for a US company.

Mira Murati takes over after Altman’s hectic eviction

Altman was “not consistently honest in his communications with the board,” it says. This hindered the council’s work. “The Council no longer has confidence in its ability to lead OpenAI.”

The expulsion was so rough, came so suddenly before the market closed on Friday, that it gave the impression that Altman’s opponents were acting in a hurry – or even in fear of being associated with him for any longer.

The interim boss will be 35-year-old Mira Murati, previously the company’s head of technology. The engineer is supposed to be looking for a new boss.

Altman’s statement about the expulsion is not very enlightening: he “loved” his time at OpenAI, he wrote on X. He will comment on what he will do next.

Supervisory board chairman Greg Brockmann also has to go

In addition to Altman, another co-founder is leaving OpenAI. Board Chairman Greg Brockman was forced to leave the council. He then resigned.

Brockman at least gave his side of the story on X: “Sam and I are shocked and saddened by the Council’s actions today.” Both “are still trying to figure out exactly what happened.”

Brockman writes: Thursday night Altman received a message from Ilya Sutskever, the OpenAI co-founder who sits on the council like Altman, Brockman and three other people. Sutskever is considered the brilliant mind behind many of OpenAI’s technical milestones. Brockman writes that he invited Sam to a video call. The entire council – except for him, Brockman – was gathered there. Sutskever told Altman he was fired. Brockman found out in another call from Sutskever that he would be kicked off the supervisory board.

Only one of the core OpenAI team remains

So Sutskever played a crucial role in the departure of Altman and Brockman. The three are part of OpenAI’s core team, in contrast to the other council members: the tech companies Adam D’Angelo and Tasha McCauley as well as the technology researcher Helen Toner are seen more as advisors than as power players.

The industry website The Information According to some OpenAI employees, they speculate that Sutskever accused Altman of putting the commercialization of Chat-GPT above the organization’s code of ethics. Sutskever belongs to the faction that is afraid of a technical “superintelligence” and therefore only wants to introduce AI cautiously. However, Altman also often warned about his own technology and compared himself to Robert Oppenheimer, the father of the atomic bomb.

However, whether it is really a “coup,” like some employees whisper, or whether Altman was actually guilty of something, was still unclear on Saturday afternoon. It is not known what things Altman is said to have kept secret from the supervisors. The sensational company’s figures may have been worse than what appeared to the outside world. OpenAI raised money through subscriptions to use its chatbot Chat-GPT.

OpenAI was founded in 2015 by Tesla boss Elon Musk, Altman and the computer scientists Brockman and Sutskever. Musk left in 2018 amid controversy. Initially, OpenAI operated as a non-profit with no intention of making a profit. Today it consists of a for-profit company and a non-profit part. The supervisory board is part of the non-profit and is intended to ensure that the original mission does not lose focus: to develop safe AI for the world.

The council had recently received new powers, including one that OpenAI considers particularly important: the power to decide when the company’s AI crossed the threshold to “artificial general intelligence” (AGI). has. At the theoretically assumed AGI level, AI can solve many different tasks at least as well as a human; the software could then abstract things really well. AGI is something like the holy grail of AI research. It is said to be a precursor to “superintelligence”.

Not the first spontaneous expulsion for Sam Altman

It is clear that Altman’s expulsion damages trust in OpenAI’s technology. He was the face of the company, and major investor Microsoft has made itself dependent on him to a certain extent. It’s not just the harsh choice of words that suggests that the expulsion and its communication were chaotic. The US website According to Axios There are indications that OpenAI informed its major investor Microsoft just a minute before the public that Altman had to go. Interim boss Murati according to Microsoft boss Satya Nadella remains fully behind OpenAI.

In addition, OpenAI wanted to sell shares of its employees worth $86 billion to get fresh money. With annual sales of a billion dollars, this was an astronomical valuation that can only be explained by the hype surrounding the company. It is questionable whether investors are still willing to pay so much money after the irritating expulsion.

US reporter Eric Newcomer points pointed out that there had already been a case in the past in which Altman had had a falling out with his employer under mysterious circumstances. It’s about the “incubator” called Y Combinator, the company is probably the best-known launching pad for start-ups in Silicon Valley and supported, among others, AirBnB and Dropbox in their early phase. After five years as president, Altman first became chairman of the supervisory board in 2019, then an “advisor” within weeks – and then suddenly had nothing to do with Y Combinator at all. Reporter Newcomer says a source told him Altman was simply absent too often.

source site