今PON,OpenAI的灵魂人物伊利亚·苏茨克维(Ilya Sutskever)ࣩ ඃ布成立新公司SSI(超级安全智能Safe SuperIntel hose)。
And just half a year ago, in OpenAI's "Gong Dou", Ilya, as the main promoter, kicked CEO Sam Altman (Sam Altman) out of the company, which was controversial. Many netizens complained about Altman.
Nowadays, roles are reversed. The AI security company established by Ilya has won the support of netizens, but Altman has become a "hypocrite" in the mouth of netizens. Why is that?
OpenAI：假公益，真赚钱？
This "denunciation" of Altman stems from a "message that OpenAI will be turned into a for-profit entity." Altman recently told shareholders that he intends to completely turn OpenAI into a for-profit entity and no longer controlled by a non-profit organization.
In fact, every time OpenAI has a new development, the statement about a non-profit intelligence research organization will be dragged out and discussed by many netizens. The main focus of the discussion was that OpenAI is "for-profit" in the name of "non-profit".
In fact, long before ChatGPT entered the public eye, Altman pushed for the establishment of a restricted for-profit entity "OpenAI LP" to solve this problem. The company is fully controlled by the former OpenAI (a non-profit organization) but can operate for profit.
At the same time, Altman also attracted many LPs (limited partners) to invest, and Microsoft was one of them. With the company's profit-making nature, limited partners can receive annual profit dividends. And the "limited" here refers to OpenAI's dividends with a return of only 100 times the upper limit. In other words, there is a fixed amount of money earned, and after it is reached, you cannot continue to make profits.
In Altman's view, such an organizational structure limits his footsteps. In addition, in this round of AI competition, "heaping data, heaping talents, and heaping computing power" has become an open secret to winning, and all three need "money" to solve.
There are only two ways to grow in the business world: one is to rely on the power of capital to achieve scale expansion; The other is to rely on the strength of the organization to achieve continuous growth. Obviously, in order for OpenAI to continue to occupy a high position in this round of AI competition, Altman intends to "take a two-pronged approach" - while bringing in new directors to "sit" and pushing OpenAI to turn into a for-profit organization.
After becoming a for-profit organization, the "100x return" limit no longer exists, and such a return incentive can further expand the range of partners. At the same time, if it moves towards an IPO (public offering), this star company, which can triple its valuation in half a year, will attract more investors to enter the market. And the current valuation of $86 billion (about 624.4 billion yuan) also has the potential to bring great wealth to Altman himself.
However, as soon as this news came out, public opinion was not so optimistic. Netizens have accused Altman, believing that he "shouted a vision for the benefit of all mankind, and used data and materials to train AI models at will in the name of a non-profit organization, and turned around and prepared for an IPO."
In the eyes of netizens, his mouth is full of isms, and his heart is full of business. What's more, he went to the comment section of Altman's account and scolded him as a "liar". Columnist Edward Zitron wrote an article directly on his blog titled "Sam Altman Is Full Of Shit." In the foreign community Reddit, some netizens specially built a topic area called "Why is Altman so annoying?" This topic triggered hundreds of comments from netizens.
Netizens replied under the topic
Only half a year ago, when Altman was forced to leave OpenAI, netizens also defended it, believing that OpenAI can come to this day without Altman's efforts, he is a "genius-level strategist". However, now he has become a "hypocrite" and a "white lotus". What made netizens' attitudes take a 180-degree turn?
When will public opinion reverse?
The timeline is pulled back to March 2023, and OpenAI released GPT-4, announcing that AI has entered a new era. Soon after, Elon Musk began attacking OpenAI on social media around issues such as the potential dangers of AGI. Altman at the time chose to participate in an interview to respond to this question.
In the interview, Altman specifically mentioned one of Musk's previous interviews. In the interview at the time, Musk was experiencing the development difficulties of SpaceX, and the host asked Musk on the show: "Many space pioneers are lashing out at you and against you, how do you feel now?" Musk replied with red eyes, "It's very uncomfortable." They are my heroes. ”
Altman put himself in the role of Musk at the time, saying:
"I look up to Musk, who was undoubtedly my hero growing up. I know he's felt the pain of not being understood, so I'm especially sad when he does it (attacking OpenAI, appealing, etc.). ”
"I wish he could come and see how much effort we're putting into our vision."
The interview has been viewed 6.43 million times on YouTube alone, and has also caused a lot of discussion on Twitter. Altman's empathy made netizens begin to stand for him, thinking that Musk was just "red-eyed" OpenAI's current achievements.
Support for Altman didn't last long. By the end of last year, when OpenAI's "Gong Dou Drama" arrived, many netizens began to have questions about Altman.
The initiator of this "palace fight" is Ilya, the former chief scientist of OpenAI. He is a recognized scientist with technical ideals, and he is also the soul of OpenAI. The argument that persuaded the board to fire Altman was that Ilya "saw the capabilities of AI" and decided that it was not appropriate to expose the public to AI at this time. This is very different from Altman's aggressive commercialization.
The OpenAI team took a group photo, with Ilya standing in the middle
In this "palace fight", Altman resigned and reinstated. Some OpenAI employees jumped to Microsoft and followed Altman back. Nothing seems to have changed, but since then, Ilya has never been involved in any company dynamics in public. Another update is that it recently announced that it has withdrawn from OpenAI and established a new AI security company, SSI (Safe SuperIntelligence).
And "what did Ilya see?" (What did Ilya see)" has also become a question in the minds of netizens. Under Ilya's post announcing the establishment of a new company, there are still netizens who keep asking this question. Search for related terms on Twitter, and you can also see thousands of posts asked by netizens. There are also quite a few memes circulating on the Internet.
The key figure that made OpenAI truly famous was actually Elijah, but after a failed infighting, Elijah was kicked out of the table, Altman returned to power, and the balance between AI safety and commercialization was completely broken. But Altman doesn't intend to stop there.
Recently, OpenAI adjusted its board seats. Among the four new directors, Paul Nakasone, the former head of the US National Security Agency, has come to take charge of security issues. The remaining three are Dr. Sue Desmond-Herman. Sue Desmond-Hellmann, Nicole Seligman and Fidji Simo are all directors with extensive IPO experience, many of whom have also worked on Facebook, the head social networking app. Invite these business tycoons, Altman seems to want to make the fire of OpenAI's commercialization burn more vigorously.
With the "loosening" of the organization and the blessing of capital, the "two-pronged" Altman completely let go of his hands and feet and began the road of commercialization. Netizens are becoming more and more worried, especially when OpenAI announced a partnership with Apple.
Uncertainty is Altman
The controversy surrounding Altman has affected the cooperation between OpenAI and Apple to some extent.
In an exclusive interview after the WWDC2024 conference, Cook was asked, "Have you been upset or even reconsidered recently about something that has raised questions about Altman's judgment and the credibility of the product?" ”
Cook's answer was very rational, and the reason for choosing OpenAI to work with him was simply "now that they have the best model".
Cook added: "Obviously, we're not going to rely on one person or one company forever. We're also working with others. It is reported that Apple executive Craig Federighi (Craig Federighi) hinted that an agreement with OpenAI's "fierce rival" - Google Gemini may be reached in the future.
At the time of the cooperation with OpenAI, Apple, which is known for its privacy and security, also carried out some subtle processing: for requests that cannot be handled by the IOS system, users need to make it clear that ChatGPT is voluntarily selected before calling OpenAI. The move is a clear demarcation of the boundaries between user privacy and security, and although it is a partnership, Apple intends to build a security wall between OpenAI and the iOS ecosystem.
On the same day at the WWDC scene, Altman, who was present to support, never had a chance to share on stage.
In the early days of OpenAI, Altman, the young president of a well-known venture capital company in the United States, was the most powerful spokesperson for OpenAI. He used his star effect to bring great impetus to the development of OpenAI. The public relations battle with Musk also made many netizens begin to believe that this is a "dragon slayer" in the AI world.
However, with the turn of word-of-mouth, various details of OpenAI are also constantly amplified and scrutinized on the Internet. "ClosedAI" has become a black name for OpenAI. Netizens used "Closed" to ridicule OpenAI's lack of open source, opacity, irresponsible grabbing of data, and abuse of non-profit privileges.
In this AI drama, Sam Altman's personal IP has become a symbol of strong binding to OpenAI.
In the first act, OpenAI, as a start-up, smoothly survives every question in a weak posture in the game with large companies, and the audience can also see the sincerity and enthusiasm of the teenager to change the world.
However, in the second act, as the plot progresses, the "dragon slayer boy" seems to have another purpose in the eyes of the audience. Once the seeds of doubt in the hearts of the audience are planted, the strong binding of personal IP and the company is no longer a contributing factor, but has become a huge hidden danger for the company's development, and the successive appearance of words such as "liar" and "ClosedAI" proves this.
The drama of AI continues, can OpenAI continue to play its first-mover advantage and occupy the top spot in the market? And how should Altman face public skepticism? I believe that the more exciting is always the next scene.