Nvidia Workers Warned CEO Jensen Huang That AI Might Be Harmful to Minorities: Report

• Two Nvidia workers met CEO Jensen Huang in 2020 to warn that AI could negatively impact minorities.
• Both were unsatisfied with his response, according to Bloomberg, and quit the company soon afterward.
• One posted on LinkedIn the meeting was the “most devastating 45 minutes of my professional career.”

Two Nvidia employees warned CEO Jensen Huang that AI might be harmful to minorities — and one was so dissatisfied with his response that she quit shortly after, according to a Bloomberg report.

Nvidia employees Masheika Allgood and Alexander Tsado met with Nvidia boss Huang in 2020 to raise concerns from the company’s employees about the dangers of AI, especially to minorities.

Both came away unsatisfied, according to Bloomberg, and left the company shortly afterward. Allgood said she left because Nvidia “wasn’t willing to lead in an area that was very important to me.”

Tsado, a former product marketing manager at Nvidia, told Bloomberg that he wanted Huang to understand the importance of the issue.

“We’re building these tools and I’m looking at them and I’m thinking, this is not going to work for me because I’m Black,” he said.

In the meeting, which Allgood described as the “most devastating 45 minutes of my professional life” in a post on LinkedIn, the two former presidents of the company’s Black employees group warned that the unforeseen impacts of the new technology would be disproportionately felt by minority groups. The duo spent a year collecting employee concerns about AI.

Allgood told Bloomberg that she raised concerns that AI facial recognition systems used in self-driving car technology could pose a threat to minorities, with studies suggesting that some facial recognition technologies have more difficulty identifying people of color.

She said Huang pushed back on this, replying that self-driving vehicles could be tested on highways where they would pose less of a threat to pedestrians.

Self-driving technology has come under increased scrutiny in recent months following a grisly accident involving a Cruise robotaxi in San Francisco.

In the weeks after the crash, which led to Cruise recalling its entire fleet of autonomous vehicles, reports suggested that the driverless car company’s technology had difficulty identifying children and large holes in the past.

Similar concerns have been raised around AI and minorities. The biased nature of many of the huge datasets that AI models are trained on has meant that some have shown discriminatory behavior.

In 2015, Google faced backlash after its image recognition systems were found to classify Black people as “Gorillas” — an issue it fixed by blocking it from identifying gorillas altogether.

More recently, AI image generators such as OpenAI’s DALL-E have been found to perpetuate harmful stereotypes when prompted for images of “African workers.”

Nvidia is a vital part of the AI arms race

Nvidia has seen its share price soar as a result of the AI revolution, with tech companies clamoring for its graphics processing units (GPUs) that are crucial for training AI models.

A company spokesperson told Bloomberg that the chipmaker had undertaken substantial work to make its AI products safe and inclusive since Allgood and Tsado departed.

Huang did give the green light to some of the proposals suggested by Allgood, according to internal emails viewed by Bloomberg, and the company has now established an AI & Legal Ethics program.

However, the chipmaker continues to lag behind other tech companies in diversity. 7.8% of its US workforce is Black, Hispanic, or Native American, according to data compiled by Bloomberg, compared to 18% at rival Intel, for example.

Nvidia did not immediately respond to a request for comment from Business Insider, made outside normal working hours.