MIT Technology Review · Jan 21, 2025
Alvin Wang Graylin and Paul Triolo argue in MIT Technology Review for US-China AI cooperation
Markdown · JSON · Original source
Published January 21, 2025, Alvin Wang Graylin and Paul Triolo argue that the US-China AI rivalry is a dangerous zero-sum framing, not a path to lasting victory. They say the capability gap between leading US and Chinese models has essentially disappeared, while Chinese models may achieve near-equivalent results using a small fraction of Western labs’ compute. In their account, US have encouraged China to pursue self-sufficiency, even as both sides tentatively engage on AI safety and governance. The authors locate the greater existential threat in advanced AI being misused by bad actors, whom they say may face fewer restraints than nation-states. They call for the US and China to cooperate on reducing misuse and governing advanced models, rather than erecting further barriers. Their conclusion is that continued rivalry risks global stability and scientific progress, while collaboration could help realize AI’s potential in areas including health care, education and scientific discovery. They urge policymakers to shift from confrontation toward shared governance and research. Whether that cooperation can take hold amid rising tensions remains uncertain.
Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.
Original source
MIT Technology Review · Article
Source details
Original source date: Jan 21, 2025 (2025-01-21T17:42:15Z).
Date basis: Date from the publisher’s own records, through its API.
Added to Escape Meridian: 2026-11-16T00:43:26.443Z. First observed: 2026-11-02T06:46:03.241Z. Retrieved: 2026-11-02T06:47:33.896Z.
Referenced by
Recorded citations and public file attachments. A reference alone does not establish agreement or influence.
No public references recorded here yet.
Limits
Text as retrieved on November 2, 2026; originally published January 21, 2025. The article makes broad claims about model capabilities, , AI misuse and possible benefits, but this text does not provide supporting studies or measurements for those claims. The source presents collaboration and confrontation as competing paths; it does not establish what would enable cooperation to persist amid geopolitical tensions.
Open question
What would enable the US and China to sustain cooperation on AI safety and research while their rivalry intensifies?
Source-stated
Graylin and Triolo argue that the US-China AI rivalry cannot produce long-term winners and risks harming global stability and scientific progress.
there will not and cannot be any long-term winners if the intense competition continues on its current path. Instead, the consequences could be severe—undermining global stability, stalling scientific progress
Source-stated
The authors say the capability gap between leading US and Chinese models has essentially disappeared, and that Chinese models may achieve near-equivalent results with far less compute.
the capability gap between leading US and Chinese models has essentially disappeared, and in one important way the Chinese models may now have an advantage: They are able to achieve near equivalent results while using only a small fraction of the compute resources
Source-stated
The authors argue that export controls have prompted China to accelerate self-sufficiency and indigenous innovation, causing US efforts to backfire.
China has responded by accelerating its efforts toward self-sufficiency and indigenous innovation, which is causing US efforts to backfire.
Source-stated
Graylin and Triolo identify AI misuse by bad actors as the central existential threat and call for US-China cooperation to address it.
the real existential threat ahead is not from China, but from the weaponization of advanced AI by bad actors and rogue groups who seek to create broad harms, gain wealth, or destabilize society.
Source-stated
The authors conclude that rivalry risks global harm, while collaboration offers the potential for a prosperous and stable future.
We can proceed down a path of confrontation that will almost certainly lead to mutual harm, or we can pivot toward collaboration, which offers the potential for a prosperous and stable future for all.
Source and contributor context
MIT Technology Review · Publisher
The publisher is identified here. No individual author has been inferred.
Source media
Media retrieval status: unchecked.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
Export controls
Government rules that limit which goods and technologies can be sold abroad, and to whom.
Editorial provenance
Version 1. Publication ID: 946a055a-ce2d-b6b0-1248-d301ce6ab007.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: passed by anthropic/claude-opus-5.5 on 2026-11-16T00:39:38.635Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: bridge_admitted_by_model, second_opinion_categorical.
Observation: b8a5e58edb16c03c7d28efe0d095b5a891f0be799615229818d791415aaedcb1. Source text hash: 36a3f8eb926f5e5c0266778ede6cc4eb11a436ef38f6e9c8fd5ea41c77ac59f1.
Revision and source-check history
Version 1 · 2026-11-16T00:43:26.443Z
Alvin Wang Graylin and Paul Triolo argue in MIT Technology Review for US-China AI cooperation
Published January 21, 2025, Alvin Wang Graylin and Paul Triolo argue that the US-China AI rivalry is a dangerous zero-sum framing, not a path to lasting victory. They say the capability gap between leading US and Chinese models has essentially disappeared, while Chinese models may achieve near-equivalent results using a small fraction of Western labs’ compute. In their account, US have encouraged China to pursue self-sufficiency, even as both sides tentatively engage on AI safety and governance. The authors locate the greater existential threat in advanced AI being misused by bad actors, whom they say may face fewer restraints than nation-states. They call for the US and China to cooperate on reducing misuse and governing advanced models, rather than erecting further barriers. Their conclusion is that continued rivalry risks global stability and scientific progress, while collaboration could help realize AI’s potential in areas including health care, education and scientific discovery. They urge policymakers to shift from confrontation toward shared governance and research. Whether that cooperation can take hold amid rising tensions remains uncertain.
Original source
MIT Technology Review · Article
Source details
Original source date: Jan 21, 2025 (2025-01-21T17:42:15Z).
Date basis: Date from the publisher’s own records, through its API.
Added to Escape Meridian: 2026-11-16T00:43:26.443Z. First observed: 2026-11-02T06:46:03.241Z. Retrieved: 2026-11-02T06:47:33.896Z.
Limits
Text as retrieved on November 2, 2026; originally published January 21, 2025. The article makes broad claims about model capabilities, , AI misuse and possible benefits, but this text does not provide supporting studies or measurements for those claims. The source presents collaboration and confrontation as competing paths; it does not establish what would enable cooperation to persist amid geopolitical tensions.
Open question
What would enable the US and China to sustain cooperation on AI safety and research while their rivalry intensifies?
Source-stated
Graylin and Triolo argue that the US-China AI rivalry cannot produce long-term winners and risks harming global stability and scientific progress.
there will not and cannot be any long-term winners if the intense competition continues on its current path. Instead, the consequences could be severe—undermining global stability, stalling scientific progress
Source-stated
The authors say the capability gap between leading US and Chinese models has essentially disappeared, and that Chinese models may achieve near-equivalent results with far less compute.
the capability gap between leading US and Chinese models has essentially disappeared, and in one important way the Chinese models may now have an advantage: They are able to achieve near equivalent results while using only a small fraction of the compute resources
Source-stated
The authors argue that export controls have prompted China to accelerate self-sufficiency and indigenous innovation, causing US efforts to backfire.
China has responded by accelerating its efforts toward self-sufficiency and indigenous innovation, which is causing US efforts to backfire.
Source-stated
Graylin and Triolo identify AI misuse by bad actors as the central existential threat and call for US-China cooperation to address it.
the real existential threat ahead is not from China, but from the weaponization of advanced AI by bad actors and rogue groups who seek to create broad harms, gain wealth, or destabilize society.
Source-stated
The authors conclude that rivalry risks global harm, while collaboration offers the potential for a prosperous and stable future.
We can proceed down a path of confrontation that will almost certainly lead to mutual harm, or we can pivot toward collaboration, which offers the potential for a prosperous and stable future for all.
Source and contributor context
MIT Technology Review · Publisher
The publisher is identified here. No individual author has been inferred.
Source media
Media retrieval status: unchecked.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
Export controls
Government rules that limit which goods and technologies can be sold abroad, and to whom.
Editorial provenance
Version 1. Publication ID: 946a055a-ce2d-b6b0-1248-d301ce6ab007.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: passed by anthropic/claude-opus-5.5 on 2026-11-16T00:39:38.635Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: bridge_admitted_by_model, second_opinion_categorical.
Observation: b8a5e58edb16c03c7d28efe0d095b5a891f0be799615229818d791415aaedcb1. Source text hash: 36a3f8eb926f5e5c0266778ede6cc4eb11a436ef38f6e9c8fd5ea41c77ac59f1.
passed · 2026-11-16T00:39:38.635Z · anthropic/claude-opus-5.5
Publication: 946a055a-ce2d-b6b0-1248-d301ce6ab007
attribution — pass: The reading attributes the op-ed to Alvin Wang Graylin and Paul Triolo in MIT Technology Review, matching the byline and author bios. It is a single argumentative piece, not a listing, and the reading frames its claims as the authors' assertions.
Alvin Wang Graylin is a technology executive, author, investor, and pioneer
summary — pass: The title, summary and question accurately convey the authors' thesis: no long-term winners, a closing capability gap, export-control backfire, misuse as the greater threat, and a call for cooperation. The language is accessible, and export controls is defined beside the reading. Compute is used without a gloss, but its meaning is clear enough from context. The question addresses a real open issue the source leaves unresolved. Omitting the seven recommendations and the Raimondo quote does not distort the argument.
it is incumbent on the US and China as global leaders in developing AI technology to jointly identify and mitigate such threats
date — pass: The reading's January 21, 2025 matches the UTC day of originalPublished, 2025-01-21T17:42:15Z. The Most Popular sidebar includes later items, such as one ten years after AlphaGo, but the reading does not use them. modifiedAt is one second after publication. Forecasts are presented as the authors' views, not outcomes, and the retrieval date is disclosed.
Ten years after AlphaGo’s match against Go champion Lee Sedol
support — pass: Each claim is supported by a verbatim excerpt from the article body and is framed as the authors' argument. The benefits mentioned (health care, education, scientific discovery) appear in the source. The limitations correctly note that the authors provide no supporting measurements and do not independently verify their capability-gap or other claims.
AI will help us accelerate scientific discovery, develop new drugs, extend our health span
media — pass: No media is selected, and the evidence supplies no verified media candidates, so there is no media identity or relevance error.
Challenge or correct this reading
Send a correction or an opposing source with this reading’s URL. Disagreement and uncertainty can remain unresolved.
Government rules that limit which goods and technologies can be sold abroad, and to whom.
Definition referenceGovernment rules that limit which goods and technologies can be sold abroad, and to whom.
Definition referenceGovernment rules that limit which goods and technologies can be sold abroad, and to whom.
Definition referenceGovernment rules that limit which goods and technologies can be sold abroad, and to whom.
Definition reference