Microsoft Eliminates Ethics and Society Team Responsible for AI Tool Development

Microsoft has recently laid off its ethics and society team, which was responsible for ensuring the company’s AI principles were closely linked to product design. The move came as part of the company’s larger effort to lay off 10,000 employees across the organization. The team’s elimination leaves Microsoft without a dedicated group of professionals to oversee its AI initiatives, especially as it continues to make AI tools more accessible to mainstream users.

While Microsoft still maintains an Office of Responsible AI, tasked with creating rules and principles to guide the company’s AI efforts, the ethics and society team played a critical role in ensuring that those principles were reflected in the design of the products that shipped.

The team’s primary responsibility was to create rules in areas where none existed, a task that proved increasingly difficult as Microsoft scaled back its workforce.

The ethics and society team was at its largest in 2020, with around 30 employees, including designers, engineers, and philosophers. However, in October, the team was downsized to just seven individuals as part of a company-wide reorganization.

In a meeting with the team, John Montgomery, corporate vice president of AI, informed employees that the pressure to move OpenAI models and other AI tools into customers’ hands at a faster pace was high. This pressure was the primary reason for the team’s elimination.

Despite this, Microsoft insists that it remains committed to responsible AI development. The company has increased the number of people working on its product teams and within the Office of Responsible AI to ensure its AI principles are put into practice. Microsoft recognizes and appreciates the work done by the ethics and society team and the critical role they played in promoting responsible AI development.

Moving forward, Microsoft plans to devolve the team’s abilities and responsibilities to individual product teams building services and software, making it challenging to implement their ambitious plans. While the central hub that did some of the work is no longer operational, the Office of Responsible AI remains active and committed to creating rules and principles to govern the company’s AI initiatives.