Former OpenAI employees sent a letter to the board of directors to expose Altman's misconduct, and Musk forwarded: The problem is worth investigating! The company's vice president quickly issued a defense
Reporter: Cai Ding Editor: Lan Suying
The OpenAI farce, which began last Friday (November 17) PST, not only did not end, but intensified.
On the morning of Wednesday (November 22), Beijing time, Musk reposted a GitHub link on X (formerly Twitter), saying that "this letter about OpenAI was just sent to me, and these questions (in the letter) seem to be worth investigating."
Musk's Retweet Image source: X
Although the GitHub link has shown "404" and cannot be opened, there are still netizens on the X platform who have published the content of the letter. Judging from the full text of the letter, a person claiming to be a former OpenAI employee attacked former OpenAI CEO Altman and former board chairman Greg Brockman, citing a series of misconduct by the two during their tenure at OpenAI, and calling on OpenAI's board to conduct a full investigation. As of press time, OpenAI, Altman, and Brockman have not given any response to the content of the letter.
Former employees accused Altman and Brockman of fraud and manipulation
In the letter, which was addressed to OpenAI's board of directors, an employee who claimed to have worked at OpenAI said the letter was written to express their deep concern about recent events at the company, particularly allegations of misconduct against Altman.
"We are former employees who left during a period of significant upheaval and change in the company. As you can see with your own eyes, what happens when someone dares to challenge Altman, you might understand why many of us choose to remain silent out of fear of reprisals. But now, we can no longer remain silent. The author of the letter wrote.
The authors of the letter believe that it is incumbent upon OpenAI's board of directors to fully investigate these allegations and take appropriate action. The authors strongly recommend that OpenAI's board of directors first expand the scope of the investigation into Emmett Hill (OpenAI's new interim CEO) to include a review of Altman's conduct since August 2018, which was the beginning of OpenAI's transformation from a nonprofit to a for-profit entity; Former OpenAI employees who are on sick leave or have been laid off have launched a call for a public call for private statements and, finally, to protect the identities of those who come forward and ensure that they are not subject to retaliation or other forms of harm.
The authors of the letter said they believe that many OpenAI employees were forced to leave in order to facilitate the company's shift to a profit model. This is evident from the company's 50% employee turnover rate between January 2018 and July 2020.
"During our time at OpenAI, we witnessed the disturbing deception and manipulation that Altman and Brockman exhibited in pursuit of Artificial General Intelligence (AGI). However, their approach raises serious doubts about their true intentions and whether they truly put the interests of all humanity first. The letter reads.
Initially, many of them were hopeful about OpenAI's mission and chose to trust Altman and Brockman, the authors say. But as the actions of Altman and others have become increasingly worrying, those who dare to voice dissenting opinions have been silenced or forced to leave. This systematic crackdown on dissent has created an atmosphere of fear and oppression, effectively stifling any substantive discussion of the ethical implications of OpenAI's work.
The author also cites a number of cases of dishonesty and "manipulative behavior" by Altman and Brockman in the letter. For example, Altman asked OpenAI researchers to delay reporting on the progress of certain "secret" research programs that were eventually terminated for failing to produce quick results, and those who questioned this were considered "culturally incompatible" and even fired, some even on the eve of Thanksgiving in 2019. In another example, Altman has secretly and frequently used OpenAI's nonprofit resources to advance his personal goals, especially out of resentment after breaking with Musk.
The authors bluntly say that OpenAI's governance structure was carefully designed by Altman and Brockman to prevent employees from participating in the regulation of profitable operations, mainly because they have an inherent conflict of interest with Altman and Brockman. This opaque structure also allows Altman and Brockman to operate with impunity and avoid accountability.
At the end of the letter, the authors stated, "We urge the OpenAI Board of Directors to take a firm stand on these unethical practices and to initiate an independent investigation into Altman and Brockman's actions." We strongly believe that OpenAI's mission is too important to be compromised by the personal goals of a few. ”
OpenAI's VP Stands Up for the Former Leadership "Platform"
The reporter of "Daily Economic News" noticed that after the content of the above letter was made public by Musk and other "big V", Peter Welinder, a user who was certified as the "vice president of OpenAI products" by the X platform, immediately came out to "stand up" for Altman.
Peter Welinder said, "I've been with OpenAI for seven years now. During this time, I also met with 'former employees who expressed concerns'. But I think their claims lack an understanding of the whole picture and are not mature enough. For example, we often argue about GPU allocation! Altman's role as CEO is to make decisions. Inevitably, there will always be people who are not satisfied. We need to look at these issues with a more mature perspective. ”
Image source: X
In addition, Peter Welinder said, "I've also given Altman a lot of feedback. I never felt insecure, but the complete opposite. I've seen him answer other people's judgmental questions with empathy. He always demanded that he be very thoughtful in terms of security and equality. The attacks on Altman and Brockman's personal qualities reflect more of the accuser's own problems. These allegations are baseless. I've personally given Brockman a lot of direct feedback over the years. He always listens carefully and acts accordingly. ”
Image source: X
Peter Welinder's X is certified as VP of Product at OpenAI, and Altman has followed him as well
The letter, which was published by Musk and other "big Vs", and Peter Welinder's reply, is just a microcosm of OpenAI's internal governance issues. According to the New York Times, the company's board of directors was divided and at odds before OpenAI's board abruptly fired Altman last Friday.
According to the report, before being fired, Altman and OpenAI's board of directors had been arguing for more than a year. This tension between the two sides has become even more acute as OpenAI has become the global leader in artificial intelligence with its popular chatbot ChatGPT.
Ilya Sutskever, a member of OpenAI's board of directors and chief scientist, believes that Altman is not always honest when talking to OpenAI's board. Board members are concerned that Altman is too focused on expansion, while they want to strike a balance between growth and AI security.
The New York Times report also said that the previous disagreement between Altman and OpenAI's board also suggests that building a new AI system is testing whether businessmen who want to make money from it can keep pace with researchers. The latter fear that if these so-called generative AI get out of hand, what they are developing could end up affecting employment opportunities and even posing a threat to humanity.
National Business Daily