Chatham House’s James Kynge says China has overtaken the US as the leading technology power: what the figures he cites show

Chatham House posted a 144-second video on TikTok on November 4, 2026 in which James Kynge, a senior research fellow at the institute and a former Financial Times Global China editor, says China “has now overtaken the United States to become the world’s leading technology power, and in the process it has left Europe far behind.” Kynge calls this “my personal definition” and adds: “Of course, that’s a generalization when we can’t generalize across every single technology.”
He cites ASPI research rankings and doctoral graduate numbers as evidence, then argues that China is accelerating. This post sets out what he said, using the video’s automatic captions, and compares each figure with the published sources.
What the video says
- Research: “out of 74 areas of advanced research, China is ahead of all other countries in the world in 66,” citing a tracking program run by ASPI, an Australian think tank.
- Talent: “in 2024, China graduated 77,000 PhDs in the STEM sciences, and the United States graduated 40,000.”
- Acceleration: even if the premise is accepted, “China is now accelerating ahead.” He says its telecoms network capacity is “already moving to two generations ahead of the UK,” which enables technologies such as “self driving cars,” and that its industrial automation “underpins an industrial automation revolution.”
Research output: the ASPI tracker
ASPI’s Critical Technology Tracker reported on December 1, 2025 that China led in 66 of 74 technologies, with the United States leading the other eight, based on 2020–2024 data. A March 31, 2026 update covering 2021–2025 said China published the greatest number of high-impact papers in 69 of 74 technologies and was rated at high monopoly risk in 41, up from 33. The video’s figure of 66 matches the December count.
ASPI defines high-impact papers as the 10 percent most cited. It says the tracker measures “not a country’s overall strength in critical technologies but its research performance in them,” and its March article describes the count as a proxy for future capability, not current capability. ASPI rates a technology at high monopoly risk when a country publishes more than three times as much high-impact research as its nearest competitor and is home to eight or more of the top 10 institutions. The December update named four newly added technologies at high risk: cloud and edge computing, computer vision, generative AI and grid integration. The March update also put Germany in the top five in 27 technologies (30 previously), the UK in 40 (48) and India in 56 (50).
Doctoral graduates
The 77,000 and 40,000 figures match a Georgetown University CSET data brief from August 2021. It projected, “based on current enrollment patterns,” that Chinese universities would produce more than 77,000 STEM PhD graduates a year by 2025, compared with approximately 40,000 in the United States. That is a projection for 2025, not a count for 2024. CSET counted four Chinese fields (science, engineering, agriculture and medicine) and seven US fields, and said Chinese graduates would outnumber US graduates more than three to one if international students were excluded from the US count.
The US National Science Foundation reported on March 13, 2026 that US institutions awarded 58,131 research doctorates in 2024, of which 45,929 (79%) were in science and engineering. Temporary visa holders earned 38% of the science and engineering doctorates among recipients who reported citizenship. NSF’s field definition differs from CSET’s, so the counts are not directly comparable. Lens did not find an official Chinese count for 2024 in the sources it checked.
Telecoms and automation
Officials at a State Council Information Office press conference on April 21, 2026 said China had 4.958 million 5G base stations at the end of March 2026 and had rolled out 5G-Advanced in 330 cities, according to CGTN. Lens did not find a published comparison that ranks UK and Chinese networks in “generations,” and the video does not say what the term refers to.
The International Federation of Robotics reported on September 24, 2026 that China installed 354,000 industrial robots in 2025, up 20% and 59% of global installations. The United States installed almost 38,500 (up 12%) and Germany fewer than 25,000 (down 8%). The global operating stock reached 5 million.
Europe
The video’s second claim is that China has left Europe far behind. In ASPI’s March update, the EU as a bloc led in three of 74 technologies and lost its first rank in small satellites. Italy was in the top five in 16 technologies (14 previously) and France in two (four previously), while the Max Planck Society and TU Delft each ranked in the global top 10 in five technologies, and TU Delft was first in quantum computing. In the IFR data, German robot installations fell 8% to fewer than 25,000 in 2025. The telecoms and AI Index figures Lens read did not include a Europe-wide comparison.
Other measures
Stanford’s 2026 AI Index says “the U.S.-China AI model performance gap has effectively closed,” with the top US model leading by 2.7% as of March 2026. Its chart caption gives Arena scores of 1,503 for the top US model and 1,464 for the top Chinese model. The index reports US private AI investment of $285.9 billion in 2025 against $12.4 billion for China, and says private investment “likely understates China’s total AI spending” because of government guidance funds. It says the United States hosts 5,427 data centers, more than 10 times any other country, and produces more top-tier models and higher-impact patents, while China leads in publication volume, citations, patent output and industrial robot installations.
Definitions
The sources measure different things. ASPI counts highly cited research papers across 74 technologies. The International Federation of Robotics counts robots installed in a year. The AI Index covers model performance, investment, data centers and patents, and the doctoral figures count degrees awarded or projected under different field definitions. The video does not name a single measure for “leading technology power.” Chatham House’s caption describes the video as Kynge on “China’s lead over the US in advanced technology.”
Sources and notes
Sources: Chatham House video on TikTok, posted November 4, 2026, with its English captions; ASPI The Strategist, December 1, 2025 and March 31, 2026; CSET, Georgetown University, “China Is Fast Outpacing U.S. STEM PhD Growth,” August 2021; NSF NCSES, Doctorate Recipients from U.S. Universities: 2024 (NSF 26-315), March 13, 2026; CGTN, April 21, 2026; International Federation of Robotics press release, September 24, 2026; Stanford HAI, AI Index 2026; Chatham House staff page. Notes: The quotations from the video come from the English captions TikTok attached to it, which the platform labels as automatic speech recognition, not from a published transcript. The captions end at about 2 minutes 10 seconds of the 2 minute 24 second video, so the last part was not captioned. TikTok’s caption spells the speaker’s name “Kygne”; Lens uses “Kynge” and takes his title and background from Chatham House’s staff listing as shown in search results, because the page itself returned an error (HTTP 403). ASPI’s December article gives 2020–2024 in its text and 2021–2025 in a figure caption, and its counts of 66 and eight leave no room for the four technologies it says the EU leads; Lens used the figures in the text. Lens did not read the full tracker data or the methods section of the CSET brief. The Stanford page does not define the 2.7% figure or give a release date, and the Arena scores come from its chart caption. The claim about UK telecoms networks could not be checked against a published source. Thumbnail photo: “Shenzhen Skyline from Nanshan” by Simbaxu via Wikimedia Commons, licensed under CC BY-SA 4.0, photographed in 2016; cropped. Illustrative photo; it does not show the events described.
Image: Chart by Lens from ASPI and International Federation of Robotics figures. Thumbnail photo by Simbaxu via Wikimedia Commons (CC BY-SA 4.0).

