Generative AI is learning to spy for the US military
1 min read
Summary
The US military is using generative AI tools developed by Vannevar Labs, allowing Marines to sift through foreign intelligence for local threats faster than was possible with the old manual process.
These tools provide reports that can be used for sentiment analysis, detecting the emotions and opinions expressed in text, and helping track foreign news reports in which military exercises are mentioned.
However, there is still a need for validation of sources, with the potential for AI to misinterpret information, leading to misinformed military decisions.
human analysts can also manipulate open-source intelligence, undermining its credibility.
However, the US military has said it will spend $100m in the next two years on generative AI applications, seeking to develop the technology further.
This is part of a growing trend, with Israel also using AI to sort through information and generate target lists for its war in Gaza, a practice that has been widely criticised.