Source trace. Via News points to the documents behind its reporting and shows what we drew from each — so you can check any claim. How we source

Source document

The Download: The Pentagon’s new AI plans, and next-gen nuclear reactors

View original at technologyreview.com

MIT Technology Review - Ai Research Title: The Download: The Pentagon’s new AI plans, and next-gen nuclear reactors Date: 2026-03-18 12:38 Source: https://www.technologyreview.com/2026/03/18/1134371/the-download-the-pentagons-new-ai-plans-and-next-gen-nuclear-reactors/ <p><em>This is today’s edition of </em><a href="ht…

Opening lines of the source · short snapshot — read the full document at the original

What we drew from this source

The claims Via News extracted from this document. We point to the source; we don't replace it.

AI models like Anthropic's Claude are already used to answer questions in classified settings, including for analyzing targets in Iran
60% confidence
The Pentagon wants to mass-produce the Lucas kamikaze drones it used to strike Iran
60% confidence
US officials say Anthropic can't be trusted with warfighting systems
60% confidence
The Pentagon plans to set up secure environments for generative AI companies to train military-specific versions of their models on classified data
60% confidence
OpenClaw is now the largest, most popular, the most successful open-sourced project in the history of humanity and is definitely the next ChatGPT
60% confidence
Training on classified data would embed sensitive intelligence—like surveillance reports or battlefield assessments—into the models themselves
60% confidence
An approaching wave of new nuclear reactors could introduce fresh challenges to nuclear waste management
60% confidence
Uncrewed narco subs could move more cocaine over longer distances without putting human smugglers at risk of capture
60% confidence
Allowing AI models to train on and learn from classified data is a major new development that presents unique security risks
60% confidence