Google and Pentagon sign deal for classified use of AI models

4 articles · Updated · The Information · Apr 28

Over 600 Google employees delivered a letter to CEO Sundar Pichai urging rejection of the agreement, citing misuse concerns.
The contract allows Pentagon use of Google AI for any lawful purpose, with language discouraging but not legally prohibiting use in mass surveillance or autonomous weapons without human oversight.
This marks a shift from Google's 2018 withdrawal from Project Maven, and follows similar, but more restrictive, AI deals between the Pentagon, OpenAI, and xAI amid ongoing employee opposition.

Google once quit a military AI project after protests. What changed, and can employees still influence Big Tech's ethics?

The Pentagon can use Google's AI for 'any lawful purpose.' What truly protects citizens from mass AI surveillance?

After the Pentagon blacklisted Anthropic, must AI firms now abandon ethics to secure government contracts?

As AI accelerates targeting, can 'human oversight' be more than a rubber stamp for a machine's lethal choice?

Is a new global arms race being fought not by nations, but by the AI companies they hire?