Trouble at OpenAI Again: The Exodus of Safety-Focused Employees

There seems to be trouble again at OpenAI, tas the company has been experiencing a significant departure of employees who are deeply concerned about the safety of AI. The most recent and notable departures are Ilya Sutskever and Jan Leike, the leaders of the company’s superalignment team, which is responsible for ensuring that AI remains aligned with its creators’ goals and does not pose a threat to humanity. However, they are not the only ones who have left.Ssince November, at least five more safety-conscious employees have either quit or been pushed out. The question on everyone’s mind once again is: What’s going on at OpenAI?

The Reason Behind the Departures

While some may speculate that a secret technological breakthrough is the cause of these departures, the reality may be more related to a loss of faith in the company’s leadership, particularly CEO Sam Altman.

According to sources familiar with the company, safety-minded employees have lost trust in Altman’s ability to responsibly handle the development of artificial general intelligence (AGI). This is the type of AI that some data scientists say can lead to the SkyNet/Terminator level of evolving intelligence.

“It’s a process of trust collapsing bit by bit, like dominoes falling one by one,” an anonymous insider revealed. The reluctance of employees to speak publicly is partly due to the non-disparagement provisions in the off-boarding agreements they are required to sign upon leaving. Refusing to sign such an agreement could result in the loss of millions of dollars in equity.

Daniel Kokotajlo, a former employee who joined OpenAI in 2022 with the intention of guiding the company towards the safe deployment of AI, refused to sign the offboarding agreement to maintain his ability to criticize the company.

“OpenAI is training ever-more-powerful AI systems with the goal of eventually surpassing human intelligence across the board. This could be the best thing that has ever happened to humanity, but it could also be the worst if we don’t proceed with care,” Kokotajlo stated.

He gradually lost trust in OpenAI’s leadership and their capacity to responsibly handle AGI, leading to his resignation.

The Turning Point: Altman’s Failed Ouster

The pivotal moment that led to the current wave of departures can be traced back to November when Ilya Sutskever, in collaboration with the OpenAI board, attempted to fire Sam Altman.

The board cited Altman’s lack of candor in his communications as the reason for their distrust. However, the ouster failed dramatically. Altman and his ally, company president Greg Brockman, threatening to take OpenAI’s top talent to Microsoft, effectively destroying the company unless Altman was reinstated.

Faced with this ultimatum, the board capitulated, and Altman returned with even more power and influence.

The Aftermath: Strained Relationships, Misaligned Priorities

Following the failed coup, the relationship between Sutskever and Altman appeared strained, despite their public display of friendship. Sutskever has not been seen at the OpenAI office for about six months. Reports state he has been remotely co-leading the superalignment team.

This team’s ambition to ensure a future AGI’s alignment with humanity’s goals seems disconnected from the company’s daily operations. These operations have been focused on commercializing products under Altman’s leadership.

Jan Leike, in his departure announcement, shed light on the challenges faced by the superalignment team. He mentioned that the team struggled to secure sufficient computing power for their work and felt like they were “sailing against the wind.”

Leike emphasized the need for more resources to be allocated toward preparing for the next generations of models. He wants a focus on security, monitoring, preparedness, safety, adversarial robustness, (super)alignment, confidentiality, and societal impact. He expressed concern that OpenAI was not on the right trajectory to address these critical issues.

The exodus of safety-conscious employees from OpenAI is a worrying sign for the future of AI development. When leading minds in AI safety, such as Jan Leike, express doubts about the world’s leading AI company’s direction, it should be a cause for concern for everyone. The loss of trust in Sam Altman’s leadership and the company’s apparent misalignment of priorities between commercialization and safety raise questions about OpenAI’s ability to responsibly develop and deploy AGI.

As AI continues to advance at an unprecedented pace, it is crucial that companies like OpenAI prioritize safety and alignment with human values. The departure of employees who are dedicated to this cause should serve as a wake-up call for the industry and the public at large.

Of course, it’s no secret that we at Human Driven AI are strong proponents of Generative AI and its incredible impact on marketing. However, it is essential that we demand transparency, accountability, and a commitment to ethical AI development from those at the forefront of this transformative technology. Only by proceeding with caution and care can we ensure that the potential benefits of AI are realized while mitigating the risks it poses to humanity.

If you need assistance understanding how to leverage Generative AI in your marketing, advertising, or public relations campaigns, contact us today. Custom training workshops are available. Or, schedule a session for a comprehensive AI Transformation strategic roadmap to ensure your marketing team utilizes the right GAI tech stack for your needs.