More than 550 Google employees have signed an open letter addressed to top executive Sundar Pichai regarding the use of AI technology by the US defence sector, reports the British newspaper Financial Times. They want Pichai to refuse to allow the government to use Google’s AI technology for covert military operations.
“We want AI to benefit humanity, not be used in inhumane or extremely harmful ways,” states the letter, which, according to the FT, was signed by more than 560 Google employees. “This includes lethal autonomous weapons and mass surveillance, but goes even further.”
The move follows a dispute between the US government and AI company Anthropic. The Pentagon wanted unrestricted use of Anthropic’s AI systems, which have attracted considerable attention with their advanced models. However, Anthropic stipulates that its technology not be deployed for the mass surveillance of the American population or for fully autonomous weapons.
In response, the US government took measures against Anthropic by classifying the company as a supply chain security risk. President Donald Trump ordered all federal agencies to stop using Anthropic’s artificial intelligence immediately. OpenAI had also previously received criticism from its own researchers following a deal with the US government, shortly after the ban on Anthropic. Sam Altman, the CEO of OpenAI, later apologised for this.
Initiative of DeepMind
According to the Financial Times, Google staff are responding to reports that the Company is on the verge of reaching an agreement with the US Department of Defence. These agreements would allow the AI tool Gemini to be used in covert operations without the conditions demanded by Anthropic.
“The only way to guarantee that Google is not linked to such damage is by rejecting all secret orders,” the letter states.
According to those involved, the letter was initiated by employees of DeepMind, Google’s AI lab. Two-fifths of the signatories work in the AI division and a similar proportion in the Cloud branch. Among the signatories are more than eighteen high-level executives. Approximately two-thirds of them chose to sign with their full names.
Jeff Dean, DeepMind’s chief scientist, has been the most outspoken on the issue to date. He stated in February that mass surveillance violates the Constitution and has a stifling effect on freedom of speech. Furthermore, he stood by an earlier promise from 2018 to ban lethal autonomous weapons.
Don’t be evil
Google has faced protests against military ties before. In 2018, several employees resigned, and thousands signed a petition against Project Maven, which used AI to improve drone strikes. Google did not renew that contract at the time and promised not to develop AI for weapons or surveillance. Last year, however, the company quietly abandoned this stance by removing passages from its ethical principles that had promised not to develop technology that directly harms people. Demis Hassabis, co-founder of Google DeepMind, stated at the time that the world has changed and that American tech companies have a duty to assist in national defence.
The letter to Pichai concludes that making the wrong choice now would cause irreparable damage to Google’s reputation and role in the world. Since 2000, an unofficial motto has lived at Google: “ Don’t be evil .” In the run-up to Google’s 2004 IPO, that text was included in the prospectus. It also became part of Google’s code of conduct. In 2018, with the formation of the parent company Alphabet, “Don’t be evil ” in the code of conduct was replaced by the less explicit “Do the right thing .”
