Professor @Penn @perryworldhouse. Former DASD for Force Development and Emerging Capabilities. Defense, #AI, and innovation. Always Boston. Views my own.

Philadelphia, PA
The US must learn the right lessons from the Iran War to stay ahead. Can't keep treating precise mass capabilities like they are a niche afterthought on offense or defense!
When it comes to offensive capacities, “the military gap between rich states and poor ones” has closed, writes @mchorowitz. How can the United States maintain its edge in this new era of warfare? foreignaffairs.com/iran/how-…
1
320
This is hopefully pretty good news, though I don't know the details. Designing and implementing sandboxes to test and evaluate frontier AI models is essential for their safe adoption in the IC and military. There's real work that has to be done and that costs $$$
Scoop: The National Security Agency is spending billions this year on testing AI models - far more than previously known Per a classified NSA estimate described by sources Compute costs for AI have exploded - including for taxpayers washingtonsun.com/technology…
1
11
983
Michael C. Horowitz retweeted
“The world is learning from Iran, building on what Ukraine had already made clear: the future of war lies in part in precise mass,” writes @mchorowitz. foreignaffairs.com/iran/how-…
9
33
7,716
New in @ForeignAffairs: Iran proved that precise mass is here. Cheap missiles and one-way attack fired at scale outperformed expectations, and no one can keep intercepting $35K Shaheds with $4M Patriots. The US has to build and change for this, not deny it foreignaffairs.com/iran/how-…
1
8
21
978
AI is not nuclear weapons, and thinking about it that way distorts how we think about international governance models
2
27
2,156
Michael C. Horowitz retweeted
Ahead of the May Trump-Xi summit, @Lingling_Wei previewed a possible US-China AI crisis hotline. At the time, I indicated that this was a good step - but that the question was whether anyone would pick up the phone. Now, it seems like it will be a summit deliverable.
We reported the AI hotline in May. Apparently it's still being discussed. wsj.com/world/china/u-s-and-… “The fundamental question is not whether we create a hotline—which I think has some value—but whether the Chinese will use the hotline, which historically they haven’t done,” @RushDoshi
3
5
23
6,523
Succeeding in the era of AI is about adoption, regardless of where you land on guardrails. It's adoption that will determine whether America or another country benefits the most artificial intelligence. And higher AI literacy will help in buying down risk!
To win the AI race, Washington should stop fixating on the prospect of superintelligence and start emphasizing “practical investments in the development and rapid adoption of AI,” argue @mchorowitz and @Lauren_A_Kahn. foreignaffairs.com/united-st…
5
1,043
Michael C. Horowitz retweeted
Forthcoming article in @ISQ_Jrnl! Trump and Xi are sitting down to talk AI this week. @AnthropicAI says we need to "pace the frontier"—and we need China's help to do it. Could the US and China ever agree on arms control for frontier AI? What would it take? 🧵 [link below] 1/
6
32
116
9,070
Michael C. Horowitz retweeted
.@mchorowitz and @Lauren_A_Kahn discuss the United States’ pursuit of artificial general intelligence—and argue that Washington should “treat the AI race with China like a marathon, not a sprint.” foreignaffairs.com/united-st…
1
4
9
7,449
Michael C. Horowitz retweeted
We are excited to announce our new cohort of Distinguished Visiting Fellows for 2026-27! 🌍 👏 Throughout the year, Pennstudents and faculty will be able to engage with and learn from this outstanding group of experts from around the world. Read more: perryworldhouse.upenn.edu/ne…
1
3
335
Have to be ready to compete in AI over time and adoption will be key - which also means guardrails can help when they build trusted confidence in the reliability of AI systems! @Lauren_A_Kahn & I ⬇️⬇️
.@mchorowitz and @Lauren_A_Kahn discuss the United States’ pursuit of artificial general intelligence—and argue that Washington should “treat the AI race with China like a marathon, not a sprint.” foreignaffairs.com/united-st…
2
903
Michael C. Horowitz retweeted
To win the AI race, Washington should stop fixating on the prospect of superintelligence and start emphasizing “practical investments in the development and rapid adoption of AI,” argue @mchorowitz and @Lauren_A_Kahn. foreignaffairs.com/united-st…
2
9
29
8,504
Michael C. Horowitz retweeted
New from @TheEconomist: a look at what China would face in a war over Taiwan featuring insight from PWH Director @mchorowitz on the strategy, technology, and risks shaping the Taiwan Strait. Read here: economist.com/asia/2026/09/1…
1
3
253
Michael C. Horowitz retweeted
PWH Director @mchorowitz on the Pentagon's aging networks in a quote for a @washingtonpost piece: "The risk is that adversaries gain a greater ability to disrupt U.S. information networks to get access to American military data and thus compromise national security." Link ⬇️
1
1
2
253
This is why we can't have nice things....
Replying to @mchorowitz
but mike, standards setting is boring and SALT cosplay is cool @DKThomp
2
899
International cooperation on AI will be incredibly hard even if everyone wants it because AI is a general purpose technology, and the possible economic gains create strong incentives for development and use. Think standard setting, not arms control.
5
1
27
4,041
Need to see more of this - both procurement and integration - from the US military in an age of precise mass. Legacy systems are not enough + now is the time to move! axios.com/2026/08/26/sandhil…?
1
406
"Winning" at AI will be all about adoption, both to capture advantages and mitigate risks! Check out @Lauren_A_Kahn & I on this in @ForeignAffairs last year
To win the AI race, Washington should stop fixating on the prospect of superintelligence and start emphasizing “practical investments in the development and rapid adoption of AI,” argue @mchorowitz and @Lauren_A_Kahn. foreignaffairs.com/united-st…
1
6
2,961