Andrew Ng 'Very Glad' Google Ends AI Weapons Pledge

Andrew Ng, the founder and former head of Google Brain, has expressed his support for Google's recent decision to abandon its commitment to not develop AI for weapons. "I'm really pleased that Google has shifted its position," Ng stated during a live interview with TechCrunch at the Military Veteran Startup Conference in San Francisco on Thursday evening. Just this week, Google removed a 7-year-old promise from its AI principles page, which had stated the company would not create AI for weapons or surveillance. Accompanying this change, Google released a blog post by DeepMind CEO Demis Hassabis, emphasizing the need for collaboration between companies and governments to develop AI that "supports national security." Google's original AI weapons pledge came in 2018, following the Project Maven protests, where thousands of employees objected to the company's military contracts. The protests focused on Google's role in providing AI to a military program that analyzed video footage, potentially enhancing the precision of drone strikes. Ng expressed confusion about the Project Maven protests to an audience primarily composed of veterans. "Honestly, when the Project Maven issue arose ... Many of you are out there, ready to risk your lives for our country to keep us safe," Ng remarked. "So how can an American company possibly refuse to assist our own troops who are out there fighting for us?" Although Ng was not at Google during the Project Maven protests, he significantly influenced the company's AI and neural network initiatives. Currently, Ng runs AI Fund, an AI-focused venture studio, and often discusses AI policy. Ng also expressed relief that two AI regulatory proposals — the vetoed California SB 1047 bill and Biden's overturned AI executive order — were no longer active. He had consistently argued that these measures would hinder open-source AI development in the U.S. Ng believes the true path to ensuring AI safety in America lies in maintaining technological competitiveness with China. He pointed out that AI drones would "completely transform the battlefield." He's not alone in this view among former Google executives. Eric Schmidt, the former Google CEO, now lobbies in Washington, D.C., for the U.S. to acquire AI drones to keep pace with China; his company, White Stork, might eventually provide these drones. While Ng and Schmidt advocate for military AI applications, the issue has long divided Google's workforce. Meredith Whittaker, now president of Signal, spearheaded the 2018 Maven protests while working as an AI researcher at Google. When Google pledged not to renew its Project Maven contracts, Whittaker expressed satisfaction, stating that the company "should not be in the business of war." She's not the only one to voice opposition. Geoffrey Hinton, a former Google AI researcher and Nobel laureate, has urged global governments to ban and regulate AI in weapons. Jeff Dean, a respected former Google executive and now chief scientist at DeepMind, also signed a letter against using machine learning in autonomous weapons. In recent years, Google and Amazon have faced increased scrutiny for their military engagements, including their Project Nimbus contracts with the Israeli government. Last year, employees from both companies staged sit-ins to protest Project Nimbus, under which Google and Amazon reportedly provided cloud computing services to the Israel Defense Forces. The Pentagon and militaries worldwide are increasingly eager to leverage AI, as the Department of Defense's chief AI officer previously told TechCrunch. As Google, Amazon, Microsoft, and other tech giants pour hundreds of billions into AI infrastructure, many are seeking to offset these costs through military collaborations.

Related article

Comments (18)