🇺🇸 The @WhiteHouse has just released its "AI Action Plan" — possibly its most impactful strategy yet.

After years spent watching DC fumble tech policy, this document is different.

It reads like it was written by people who understand both the technology and the stakes.

🧵 2/ The framing gets straight to the point:

"The United States is in a race to achieve global dominance in AI. Whoever has the largest AI ecosystem will set global AI standards and reap broad economic and military benefits."

No more "competing while we cooperate."

.@nvidia I love your products. You're probably the most important company in the world.

Just a quick question —

When you say you're "partnering" with companies sanctioned for building supercomputers for the Chinese military, what exactly do you mean by that?

🧵
2/ Your website lists Inspur Electronic Information Industry Co., Ltd. (IEIT Systems) as an "elite" NVIDIA partner.

IEIT is a major Chinese data center and supercomputer provider.

You say you sell them GPUs for use in their supercomputers. They advertise your GPUs in their supercomputers.

I recently returned to DC after a week spent in Taipei — a fascinating time to be on-island amid RightsCon, PLA live-fire exercises, and a seismic shift in U.S. policy toward Ukraine.

This was somehow my first visit to 🇹🇼, so I wanted to share a few high-level impressions 🧵: 1⃣ It’s one thing to read about, and quite another to experience in person:

Taiwan's democracy is an indelible feature of its society, culture, and place in the world.

I'm leaving the U.S. government after 2 years as @StateDept's main contact with the Chinese Embassy in Washington — the most challenging and rewarding experience of my life.

More soon on what's next.  But first, a few thoughts on diplomacy and America’s role in the world: 2/ I’m sad to leave what must be the most interesting job in Washington, and a team that includes some of the United States' most talented and devoted public servants.  

I am and always will be grateful to countless mentors who taught me the American way of diplomacy.

It's been about 6 months since I started a job in government.

The learning curve can be steep at first — especially if you're unfamiliar with the parlance of bureaucrats.

For the uninitiated, here's a 🧵 with some commonly used phrases and what they mean in plain English: "Tasker" — a thing you must do.

The Chinese military is buying high-end computer chips designed by 🇺🇸 companies to power its next-generation weapon systems.

Here's what @kelmgren, Ellen Lu, and I found in our latest investigation for @CSETGeorgetown — and how to stop it.

🧵 1/17

cset.georgetown.edu/publication/si… 2/ My last study on AI in 🇨🇳 analyzed 66,000 purchase records published by the People's Liberation Army (PLA).

In November, we began to scour them for any mention of “AI chips” — high-end GPUs, FPGAs, and ASICs that are useful for training AI systems.

cset.georgetown.edu/publication/ha…

The U.S. approach to economic statecraft is not cut out for strategic competition with China and Russia.

We are watching our adversaries source equipment, data, and capital from 🇺🇸 and its allies, and I am tired of admiring the problem.

🧵 1/10

washingtonpost.com/world/2022/06/… 2/ @HASCRepublicans, @HASCDemocrats, @SASCGOP, and @SASCDems should use this NDAA cycle to build the Department of Commerce into a juggernaut, expand open-source intelligence collection and analysis, and/or instantiate new, flexible export control authorities within DoD.

Succeeding in tech competition with China requires understanding the sources of power within its innovation system.

That's why @emily_sw1, @channingclee, @AnnaPuglisi838, and I dove into its expansive network of State Key Laboratories.

🧵 1/10:

cset.georgetown.edu/publication/ch… 2/ Our latest report and data visualization map nearly 500 of China’s “SKLs” across varying fields and physical locations.

It's absolutely packed with findings. But I want to highlight three of our most important conclusions about the state of science and technology in China:

Blinken outlines 3 lines of effort for U.S. China policy:
1. Invest to double-down on longstanding American strengths.
2. Align efforts with allies and partners.
3. Compete to defend our interests and build our vision for the future. Extremely pleased to see Blinken call out the crucial role that Chinese students play in advancing U.S. innovation, including the tendency of so many to stay and contribute to our economy and society.

I think it's time we acknowledged the new U.S. tactic to indirectly extend deterrence to Taiwan:

Sending a continuous stream of important political figures to publicly, physically visit the place.

🧵 1/7

asia.nikkei.com/Politics/Inter… 2/ Of course deterrence isn't the sole, or even primary, objective of this Congressional visit.

Bilateral exchanges like this have tremendous value for the U.S.-Taiwan relationship, and the international community benefits from 🇹🇼's participation in it.

The united front system remains one of China's most closely guarded secrets.

For @ASPI_org, I summarized what we know of its overseas operations:

1) advancing CCP political influence
2) exerting power over the 🇨🇳 diaspora
3) acquiring foreign technology

aspistrategist.org.au/how-chinas-uni… 2/ For a much more comprehensive overview of China's overseas united front work, I recommend:

* @boweconstrictr's 2018 report for @USCC_GOV: uscc.gov/research/china…

* @CliveCHamilton and @MareikeOhlberg's new book, "Hidden Hand" simonandschuster.com/books/Hidden-H…

⚡️ We know the Chinese military is betting big on AI. But how exactly will it use the technology?

To find out, my team @CSETGeorgetown spent a year analyzing 66,000 of the People’s Liberation Army’s purchase records.

Here is what we found (1/17): cset.georgetown.edu/publication/ha… 2/ Like any large organization, the Chinese military buys a lot of stuff: Toilet paper, staplers, ball bearings...and weapons systems, like combat UAVs, submarines, and fighter jets.

A shocking amount of that info is public—if you know where to look.

Look out below! 🌊

In my first article for @CIMSEC, I discuss China’s growing fleet of autonomous undersea vehicles (AUVs) and what it means for the U.S. Navy and partners in the Indo-Pacific.

Here are the highlights (1/8):
cimsec.org/leviathan-wake… 2/8 For years, U.S. and Chinese analysts have expected the People’s Liberation Army Navy (PLAN) to face significant disadvantage in undersea and anti-submarine warfare.

RAND’s 2015 report "War with China" talks at length about "Chinese weakness in ASW.”
rand.org/pubs/research_…

I've been thinking a lot about this distinction (state-backed troll vs amped up nationalist).

Let me reply to Victor's question and also respond to criticism (some made in bad-faith) of "A Different Kind of Army" (1/6):

jamestown.org/program/a-diff…

https://twitter.com/vshih2/status/1381903265818808323

2/6 The "20 million part-time trolls" in the report refers only to registered "network civilization volunteers" (网络文明志愿者) claimed by Communist Youth Leagues. There are other nationalists on the Chinese internet that are not organized by the Party/state.

In my latest for @ChinaBriefJT, I document how China's Communist Party raised an army of 22 million internet trolls—and how Beijing is wielding them as a weapon of foreign influence.

Here are the highlights (1/13):
jamestown.org/program/a-diff… 2/13 First we’ve got to talk about that number. It’s huge. Even larger than the 2 million trolls estimated in landmark 2013 & 2017 studies by @KingGary, @jenjpan, Margaret Roberts. How is this possible?
gking.harvard.edu/50C

Seeing a few people compare this China Daily video to Russian disinformation operations. The comparison is spot on.

Here are a five similarities to Russian coverage of the 2018 #Skripal poisonings that deserve your attention:

https://twitter.com/ChinaDaily/status/1348107266075938818

1. Step-by-step pseudo-refutation, attacking ethos, not substance.

The enumerated replies look like a bad high school debate flow chart. Take a look at this account, which constantly engaged in the same activity after the Skripal poisonings.

https://twitter.com/elenaevdokimov7/status/1046318522610016256?s=20

Who goes to work for China’s defense industry? In our latest, @emily_sw1 and I mapped talent flows between 🇨🇳 universities and defense SOEs.

🚨 We found that U.S. tech companies are indirectly aiding China's military.

Thread: (1/10) cset.georgetown.edu/research/unive… 2/10 In our report for @CSETGeorgetown, we measured the relationships between elite 🇨🇳 universities and China's defense industry by looking at their Graduate Employment Quality Reports.

Kudos to @alexjoske @ASPI_org for highlighting these sources: aspi.org.au/report/china-d…

In my first article for @ChinaBriefJT, I mapped the budget of China's united front, the collection of organizations the CCP leverages to silence political opponents, persecute religious minorities, and acquire foreign tech.

Here is what I found (1/9): jamestown.org/program/puttin… (2/9) For years, Chinese diplomats have insisted that the united front is nothing more than a benign administrative bureaucracy and accused Western analysts of overhyping its role.

But the CCP's own public budget documents belie its claims about the UF's importance and function.

This week @GeorgetownCSS we review foundational texts on Chinese strategic thinking -- Sun Tzu, Sun Bin, and others. But how heavily do these texts weigh on China's military, compared to, say, Clausewitz? Is there value in comparing 🇺🇸 and 🇨🇳 strategic culture in 2020? (1/6) Eurocentrism in IR and security studies programs has penetrated global military thinking, in ways I don't think Western academics appreciate.

Why do we assume Sun Tzu matters more to China's (modern) strategic thought than Western military strategists? (2/6)

This summer, @CSETGeorgetown has been publishing a whirlwind of papers about #AI and China's efforts to acquire it. Some projects have been months or years in the making. In case you missed them, here's a roundup of data-driven analyses I'm proud to have contributed to: 2/ On the military side of things, we wanted to know how PLA officers and defense engineers envision using AI in future warfare. It turns out the PLA is facing major hurdles in AI development: limited access to data, workforce issues, and a dearth of GPUs:
cset.georgetown.edu/research/chine…