Part of Trump's Ireland Visit · 28 stories · since Sep 3 · newest 15m ago
Pentagon uses AI to target Iran strikes as most Americans remain unawareMother Jones reports Pentagon uses Claude AI to rank targets and draft bombing justifications
1 Sep 12 · 11d ago · 22 articles · 12 posts · 8 comments · 5 sources · development 1 of 3
Mother Jones reports the Pentagon's Maven Smart System, built by Palantir, uses Anthropic's Claude AI to "semi-autonomously rank targets by strategic importance, drafting automated legal justifications for each strike along the way." The article notes the same Claude system that assists with military targeting is used by civilians for AI boyfriends and drafting texts.
“semi-autonomously rank targets by strategic importance, drafting automated legal justifications for each strike along the way”
Katie Livingstone, Military TimesCameron Stanley Chief Digital and AI Officer, Department of Defense
Pete Hegseth Secretary of DefenseJonathan Guyer Survey co-author, Institute for Global AffairsSophie Hurwitz Journalist, Mother Jones
The whole story articlespostscomments the bright band is this development · numbered dots are the others · click one to jump
Reported in the same hours no headline names this development itself — these 9 claims were published in its stretch
-
first by The Australian, 12d ago · also WSJ, Tom's Hardware
2 more headlines
-
first by Quartz, 12d ago · also Stars & Stripes
1 more headline
- Iran-linked actor used AI to build targeting guides on US Navy Stars & Stripes · 12d ago
-
first by The Times of India, 12d ago · also The Hindu
1 more headline
- Anthropic says Yemen fighters used AI to seek guided weapons The Hindu · 12d ago
-
first by The Indian Express, 11d ago · also Indian Express
-
2 outlets Adversaries Using Claude AI To Target Americans And Develop Missiles Is A Sign Of What’s To Come
first by The War Zone, 12d ago · also TWZ
-
first by Shelly Palmer, 10d ago
What people said 10 voices · verbatim
-
I
Oopsie daisies- 'its military and allied Houthis are eager to use American-built # AI technology to: target the US Navy and develop weapons…and develop surveillance, domestic tracking tools, and propaganda to use in Iran. In addition to building targeting recommendations against American naval forces as well as speeding up the development of…
-
If you know anything about how widely AI is used every day, it's actually pretty hard to discover and stop all cases of abuse (eg, distillation attacks, eliciting bio or cyber harms, unauthorized military planning and targeting) without blocking legitimate users, because attackers have come up with some really clever tricks for things like…
-
rogue states and non-state actors turning to ai for weapons development. anthropic discloses it caught some… but not all… of these efforts. implications of no governance getting clearer by the day
-
What? This is a threat report, they successfully detected and stopped the bad use. This is like saying Microsoft Word execs need to be thrown in prison because their word processor has made it so much easier for Al Quada to communicate or that McDonald's execs need to be mailed because some terrorists overheard police talking in one and then met…
-
Ahmad Shah Massoud, the minister of defense for Afghanistan, aka the Lion of Panjshir, warned us about an upcoming attack, 9/11. He was killed by a suicide bomber posing as a camera man for a news crew on 9/9/2001. The man was an icon and his picture was plastered all over Afghanistan when I was there. His son led the last of the forces against…
-
Working in kitchens you meet a lot of different Hispanic people, one time I was hanging out, off the clock and absentmindedly made a joke about the contras. Immediately remembered the dude making pizzas, “Carlos”, was from Nicaragua. I immediately started apologizing, and I think he knew I knew I fucked up.
-
So the new advertising plan is, "not only is it dangerous, but we sold it to our enemies!" Betty Crocker and Duncan Hines should get in on this hype and tell everyone they've been selling yellow cake to Iran.
-
Anthropic tech is responsible for more murder/death in the world Got it zero regulation zero accountability
-
That's a weird way to announce you have no fucking idea what you're talking about... ...but carry on.
-
Guys, read the actual report by Anthropic's Threat Intelligence team Unreliable narrator
All 3 developments of Pentagon uses AI to target Iran strikes as most Americans… →
NewswiresHacker NewsMastodonGoogle NewsRedditBluesky