News on July 2, according to "Wired" report,OpenAICEOSam·ultraman(Sam Altman) on Meta CEOMark Zuckerberg(Mark Zuckerberg) responded positively to recent aggressive “poaching” efforts in the field of AI. In a public statement to the OpenAI research team late Monday,Altman directly stated why staying at OpenAI is the only option for those who want to build artificial general intelligence (AGI), and emphasized that the company is re-evaluating the salary structure of the entire research department.
Ultraman also disapproved of Meta's poaching behavior, and pointed out that this approach may cause serious cultural hazards in the long run.
He wrote on Slack: "We have grown from a group of corner geeks to the most watched people in the technology industry (at least in the technology industry). Twitter in the AI circle is really messy, and Meta's style of behavior is also a bit uncomfortable. I have a feeling that the future will only be crazier. When I came back after being fired, I said that that incident was not the craziest thing in the history of OpenAI; and now this incident is obviously not the same."
The background of this statement is that on Monday, Meta CEO Zuckerberg issued an internal letter to all employees of the company, announcing the establishment of a superintelligence team, co-led by former Scale AI founder Wang Tao and former GitHub head Nat Friedman. Members of this new team also include many former OpenAI employees, such as Shengjia Zhao, Shuchao Bi, Jiahui Yu and Hongyu Ren. Last weekend, OpenAI chief research officer Chen Xinhan told the team that it felt like "someone broke into the house and stole something."
Altman, however, took a different stance on the brain drain in an internal letter on Monday.
He wrote: "Meta did poach some outstanding talents, but overall, they failed to poach our real core top talents and could only keep looking for people further down the recruitment list. They have been recruiting talents for a long time, and I can’t even remember how many times they wanted to ask someone from our side to be their chief scientist.I'm proud of our industry's sense of purpose, but of course, any industry is bound to have mercenaries who come here for profit. "
Altman added: "The 'dream chasers' will eventually defeat the 'profit chasers.'" He also mentioned that OpenAI is re-evaluating the compensation plan for the entire research team.“I believe that OpenAI’s stock potential is far greater than that of Meta, but I also believe that huge returns must be predicated on huge success. In my opinion, Meta’s current actions are likely to cause extremely serious corporate culture problems. We will have more information soon, but to me,It is vital to ensure that everyone is treated fairly, and pay adjustments are never made just for those who are ‘targeted’ by Meta."
Altman then further called on everyone to stay at OpenAI. He wrote: "I have never had such firm confidence in our research roadmap. We are betting on computing power like never before, and I am happy to see the results and believe we can make the best use of them. More importantly,I think we have the most special team and culture in the world. Our culture has definitely improved since we experienced insanely high growth. But we got the core right, which I don't think anyone else quite did. I believe that we have the ability to solve the current problems."
He added: "Perhaps most importantly, we really care about building general artificial intelligence the right way. Other companies see this as a means to achieve other goals, but for us, this will always be the core mission. Even if Meta moves to new hot spots in the future, or is busy guarding its social moat, we will still be here - day after day, year after year - continue to work hard and strive to do things better than anyone else. Many other trends have come and gone."
Subsequently, many OpenAI executives who had worked at Meta also spoke out on Slack to prove the superiority of OpenAI culture. One said: "They (Meta) are always changing their focus." Another wrote: "Yes, we are indeed a bit weird and unique, but that is why we can be the cradle of innovation magic." Someone else said: "OpenAI's weirdness is the most magical kind. We are inclusive and eclectic."