Anthropic and OpenAI Draw a Red Line for the Pentagon - DTNS 5215
Daily Tech News Show
Tom Merritt
4.8 • 1.5K Ratings
🗓️ 27 February 2026
⏱️ 24 minutes
🧾️ Download transcript
Summary
Block announced a 40% layoff of staff as it leans further into AI tools, and Netflix drops out of the bidding wear for Warner Discovery leaving Paramount in the driver's seat.
Starring Jason Howell and Jenn Cutter.
Show notes can be found here.
Hosted on Acast. See acast.com/privacy for more information.
Transcript
Click on a timestamp to play from that location
| 0:00.0 | This is the Daily Tech News for Friday, February 27th, 2026. |
| 0:09.8 | We tell you what you need to know, give you the important context, and help each other understand. |
| 0:14.5 | Today, Anthropic and the Pentagon are in a showdown over using AI for mass surveillance and autonomous weapons. |
| 0:21.9 | This is a meaty story. I'm Jason Howell. |
| 0:24.4 | I'm Jen Cutter. |
| 0:25.5 | Let's start with what you need to know with a big story. |
| 0:31.8 | There is a showdown of Bruin right now, happening between Anthropic and the U.S. Department of War over whether AI firms must give the Pentagon blanket access for all lawful purposes to their systems or to draw hard lines on mass domestic surveillance, fully autonomous weapons, all that stuff. |
| 0:57.0 | CEO Dario Amade laid out two contractual lines in the sand. The first of those, no AI-driven |
| 1:04.5 | dragnet surveillance of Americans. The second, no use of its technology for autonomous lethal weapons. |
| 1:12.2 | Now, the Pentagon officials have threatened to cancel its $200 million contract with Anthropic. |
| 1:19.6 | There's a deadline that they've set for an agreement. |
| 1:22.4 | 501 p.m. today, I'm not sure if that's Eastern or Pacific, but it doesn't matter sometime later today. |
| 1:29.3 | OpenAI CEO Sam Altman has also responded in a memo to staff saying that he shares Anthropics redline stance, but he proposes a different more technical method of enforcement. |
| 1:45.7 | That would require cloud-only deployments that continually keep humans in the loop, in other words, |
| 1:51.3 | as opposed to locally run systems that could be implemented inside of weapons, |
| 1:57.0 | autonomy, of course. |
| 1:58.7 | OpenAI also wants to preserve the company's ability to learn from deployments |
| 2:03.9 | and have researchers with security clearances who can advise on risks. And this goes beyond just |
| 2:12.0 | these two companies throughout the industry, more than 200 workers at Google. Yes, within Open AI and others have signed letters, |
| 2:20.1 | urging executives not to undercut Anthropic and to resist pressure from the Department of Defense or the |
| 2:26.9 | Department of War and to adopt similar red lines. And so really this just seems like, you know, the last couple of years, this has been |
| 2:36.9 | kind of the big bubbling question is like, okay, these AI models, they're smart, they're developing, |
... |
Please login to see the full transcript.
Disclaimer: The podcast and artwork embedded on this page are from Tom Merritt, and are the property of its owner and not affiliated with or endorsed by Tapesearch.
Generated transcripts are the property of Tom Merritt and are distributed freely under the Fair Use doctrine. Transcripts generated by Tapesearch are not guaranteed to be accurate.
Copyright © Tapesearch 2026.