Google Signs Contract with U.S. Department of Defense Despite Employee Opposition
NSA Adopts Anthropic's "Mythos" Model

Global artificial intelligence (AI) companies are closing the gap with the U.S. Department of Defense (DoD), seeking to enhance security and military operational capabilities, despite opposition from their own employees.

Manish Gupta, Director Introducing Google AI Gemini, Yonhap News

Manish Gupta, Director Introducing Google AI Gemini, Yonhap News

View original image

On April 27 (local time), around 600 Google employees sent an open letter to Sundar Pichai, CEO of Google, requesting that the company halt ongoing negotiations with the U.S. Department of Defense. It was reported that high-level executives, including senior managers, also participated in drafting the letter. The Google employees stated, "We do not want AI to be used in inhumane or extremely harmful ways," and emphasized, "If Google wants to avoid being implicated in the misuse of AI, it must refuse all classified work." In effect, they objected to the U.S. Department of Defense utilizing Google's generative AI models, such as Gemini, for purposes such as weapon development and improvements in military decision-making.


Despite the backlash from employees, Google is continuing to pursue collaboration with the U.S. Department of Defense. This stands in contrast to 2018, when Google withdrew from the Pentagon's drone video AI analysis initiative, "Project Maven," following employee protests. According to the Wall Street Journal (WSJ) and others on April 28, Google has signed a contract allowing Gemini to be used in classified operations for the Department of Defense. The contract states, "Both parties agree that the AI systems will not be used for large-scale surveillance or autonomous weapon development in the United States without appropriate human oversight and control, nor will they be used for such purposes." However, it also clarifies, "This agreement does not grant the right to control or refuse governmental decisions regarding lawful operations."


It appears that Google has chosen cooperation with the U.S. Department of Defense for the sake of national security advancement. A Google public sector representative, responsible for U.S. government-related work, said in a statement, "We are proud to be part of a consortium of advanced AI research institutes, technologies, and cloud companies providing AI services and infrastructure in support of national security."

Outstanding AI Model Performance Leads to Improved Relations Between U.S. Department of Defense and Anthropic

Global Artificial Intelligence Company Anthropic AP Yonhap News

Global Artificial Intelligence Company Anthropic AP Yonhap News

View original image

The U.S. Department of Defense also plans to utilize Anthropic's generative AI models to enhance national defense capabilities. On April 28, U.S. media outlet Axios reported that the Department of Defense is lifting its designation of Anthropic as a supply chain risk entity, and that the White House is drafting an executive order to allow for the adoption of Anthropic's high-performance AI model "Mythos." Relations between the U.S. government and Anthropic have been improving, as evidenced by a meeting on April 17 between White House Chief of Staff Susan Wiles and Dario Amodei, CEO of Anthropic.


Until as recently as March, the relationship between the U.S. Department of Defense and Anthropic was at a breaking point. The Department of Defense had sought to use Anthropic's generative AI models for the development of classified networks, but Anthropic refused. Anthropic maintained that it could not provide its services in areas that diverge from AI ethics, such as wartime operations. In response, U.S. President Donald Trump criticized Anthropic as a "radical left-wing company" and, in early March, designated Anthropic as a supply chain risk.



Experts believe that Anthropic's "Mythos" model served as a turning point for improving the relationship between the two parties. Concerns arose that Mythos, with its outstanding capabilities, could undermine cybersecurity systems, prompting the government to adopt a more favorable stance. It is interpreted as a strategic move to utilize Mythos in developing security and cybersecurity systems. Axios also reported that the National Security Agency (NSA), an intelligence agency under the Department of Defense, is currently using "Mythos Preview," a trial version of the Mythos model.


This content was produced with the assistance of AI translation services.

© The Asia Business Daily. All rights reserved. Unauthorized AI training and use prohibited.

Today’s Briefing