Discussing AI, technology, China, and whatever I feel like. @scientistsorg, formerly @CNASdc @ORFAmerica. Opinions are my own; not sure who else's they would be
Joined October 2014
- Tweets903
- Following775
- Followers321
- Likes17.5K
Pinned Tweet
American lab leaders are calling to "pace the frontier" of AI research as a prudent method to lower risks, but to Chinese eyes, these safety proposals look less like genuine outreach and more like American efforts to permanently cement an AI lead.
The United States can do little to change that with rhetorical tweaks, and ceding the frontier is out of the question. Instead we must use the tools we have now to promote global AI safety, create space for our AI companies to get it right, and do the work to prepare for the future ourselves.
A grand bargain may be hopelessly remote, but that doesn't mean AI safety has to be.
My work has been highly informed by some of the great analysis flying around the last few days.
watchfulpeace.substack.com/p…
Here are some immensely helpful things you may have missed:
@mattsheehan88: nytimes.com/2026/09/15/opini…
@JulianGewirtz: transformernews.ai/p/bargain…
@kyleichan: xitter.cf/kyleichan/status/21001…
@niubi: sinocism.notion.site/Global-…
Seth Center: nytimes.com/2026/09/16/opini…
@lmatsakis: wired.com/story/china-isnt-b…
@RyanFedasiuk: xitter.cf/RyanFedasiuk/status/20…
@selinaxu_: xitter.cf/selinaxu_/status/20998…
My 2 cents on China’s reaction to a coordinated slowdown in AI: rn Beijing is unlikely to agree to some sort of pause or pacing because (1) they already see themselves as being slowed down forcibly by US export controls/containment and in the position of the laggard catching up. (2) AI is one of the key “new quality productive forces” that Beijing needs to drive productivity gains and mitigate slowing economy headwinds. (3) Chinese labs aren’t calling for it either - their top priority is survival and profitability amid fierce domestic competition, and they’re generally burdened by all the regulatory compliance that already exists (to advocate for international policy of this sort is also seen as overreaching into the domain of the govt)
As I told @semafor and @jdcapelouto,
China’s top priority for AI talks with the U.S. is for more “constructive strategic stability,” i.e. assurances that there would be some floor for competition (for example, fewer US entity listings of Chinese AI firms).
semafor.com/newsletter/09/15…
I am pleased to note that last week I joined the AI team at the @scientistsorg where I will work on AI safety and security policy. I think there is much we can use from FAS's track record of dealing with new, and potentially dangerous, technologies to confront the challenges of AI where rapid progress shows no sign of abating, even as we struggle to catch up with the present.
If you work in AI policy, are interested in working in AI policy, think your expertise could be valuable to AI policy, or couldn't care less but really like my vibe, please reach out. I would love to connect (or reconnect) and learn about your work and see how I can help!
For those of you familiar with my work at my substack the Watchful Peace, my hope is to continue writing there wherever possible, but I expect the cadence to be far lower. If you have been a long time reader, I heartily thank you; if this is your first time hearing about it, feel free to read old pieces and snipe at my thoughts.
I couldn't be more excited to keep working on AI policy issues in this changing landscape and with some of the coolest new colleagues!
At long last we have created the Kobayashi Maru from the classic sci-fi story Cheat at the Kobayashi Maru and They Make You a Captain
A new Pentagon promotion test is putting would-be generals into simulated combat scenarios before boards decide who gets a star.
The Joint Warfighting Evaluation will give eligible senior officers military problems to solve, with some versions featuring an AI-powered adversary that responds to their decisions in real time.
"Right now in our current system, we have people that are very good at navigating their careers," Pentagon adviser Stu Scheller tells Fox News Digital, while stressing existing performance reports will remain the "bedrock" of promotion decisions.
The first evaluations are expected in 2027, and each service is taking a different approach.
See what future generals will face: foxnews.com/politics/pentago…
Massive fumble that an Albanian politician did not get a quote about this concept first
"The Albanians, along with the Chinese, are at the forefront of AI"
To be completely fair to Trump here, he does have his own version of pacing the frontier: delaying frontier models with somewhat random on-the-spot justifications
Probably not the best approach but he did undoubtedly give people more time to deal with Fable-related cyber risk
My entire timeline today
There seems to be some undeserved flak directed at @jacobstokes for this (really good) report so I want to just highlight his response but also note that I think we need more people thinking about this kind of risky behavior
I see people are discussing my recent report, Superpowers and AGI, here after @vince_chow1 published a piece about it in @SCMPNews. The report is long and multifaceted, so I thought I'd share some of its nuances.
If we get AGI first, how do we make it less likely that China does something like this?
Jake is clearly doing the necessary work on the flip side: what should we do if we fall behind? Is it worth doing something dangerous so they won't be the country who creates AGI?
The crisis of communications is not that we don't have a proper channel, it's that they duck our calls when we use it
September is Constitution month. What can we learn from their efforts to create new rules from scratch as we deal with massive changes to governance with the arrival of AI?
Not as much as we would like...
Me get played by Middle Eastern countries milking both sides of a high stakes geopolitical competition?
That's unpossible!
This is probably the most successful promotional account of all time
Every time @NielsHoven pops up on my feed, all I can think is "I need to make more money so I can get this software for my son so he can avoid learning to read from those critics"
