Google updates AI policy, removes ban on military and surveillance use

ReuterS

Google has revised its AI principles, removing previous commitments not to develop artificial intelligence for weapons or surveillance technologies.

The changes, first reported by The Washington Post, eliminate a section titled "applications we will not pursue," which had explicitly prohibited AI use in weapons and surveillance that violated international norms. The updated document introduces a broader focus on "responsible development and deployment," emphasizing human oversight and alignment with international law and human rights.

In a blog post, DeepMind CEO Demis Hassabis and Google executive James Manyika cited AI's evolution as a "general-purpose technology" as the reason for the update.

Google originally introduced its AI principles in 2018 after employee protests over its involvement in Project Maven, a military initiative using AI to analyze drone footage. Despite the initial backlash, the company has since resumed bidding on military contracts, including work with the U.S. Department of Defense and Israel's Defense Ministry.

Tags

Comments (0)

What is your opinion on this topic?

Leave the first comment