OpenAI has a new safety team — it's run by Sam Altman
Is OpenAI's New Safety Team Just a PR Stunt?
OpenAI has established a new safety team led by CEO Sam Altman, focusing on critical safety and security decisions for the company's projects. This move follows the departure of several key AI researchers and the dissolution of the Superalignment team, amid concerns that safety had become a secondary priority to product development. The team, which includes notable OpenAI figures and board members, will assess and develop safety processes, presenting their findings to the board for implementation decisions.
What specific responsibilities will OpenAI's new safety team have regarding the company's projects and operations?
OpenAI's newly formed safety team has been tasked with a critical role in shaping the company's approach to safety and security across its projects and operations. The team, led by CEO Sam Altman and board members Adam D’Angelo and Nicole Seligman, is responsible for making recommendations on crucial safety and security decisions6. Their initial duty involves evaluating and improving OpenAI’s existing processes and safeguards. Once this evaluation is complete, they will present their findings and recommendations to OpenAI’s board. The board will then make decisions on how to implement these recommendations, effectively influencing the future safety protocols and measures within the company. This initiative underscores OpenAI's commitment to addressing safety concerns, which have been highlighted by the recent departure of several key AI researchers and the dissolution of the Superalignment team2.
Given the composition of the safety team, how is OpenAI addressing the concerns raised by its former employees about the company's commitment to safety?
Formation of New Safety Team: OpenAI has responded to the concerns regarding safety by forming a new safety team led by CEO Sam Altman, along with board members Adam D’Angelo and Nicole Seligman34. This team is tasked with making critical safety and security decisions for OpenAI projects and operations. Their initial role is to evaluate and enhance OpenAI's current processes and safeguards.
Leadership and Decision-Making: The leaders of the new safety team, including Altman, D’Angelo, and Seligman, all hold seats on OpenAI's board. This configuration allows them to present their findings directly to the board and influence how the recommendations are implemented. This structure aims to integrate safety considerations more deeply into the governance of the company.
Addressing Underlying Concerns: Despite these measures, the setup of the new safety team might not fully address the concerns of former employees. Critics, including those who have recently left the company, might argue that having two board members and the CEO on the safety team does not represent a significant shift towards prioritizing safety, as these are the same leaders under whom previous safety concerns were allegedly sidelined34. This perception could undermine confidence in the effectiveness of the new safety initiatives, suggesting that more fundamental changes might be necessary to fully reassure stakeholders of the company's commitment to safety.