Back
MIT Technology Review
webCredibility Rating
4/5
High(4)High quality. Established institution or organization with editorial oversight and accountability.
Rating inherited from publication venue: MIT Technology Review
Published January 2025 following DeepSeek's release, this piece challenges the dominant 'winning the AI race' narrative and argues for US-China cooperation on AI safety as a strategic imperative rather than a concession.
Metadata
Importance: 52/100news articlecommentary
Summary
This MIT Technology Review article argues that the US-China AI arms race framing is counterproductive and increasingly untenable, as China's DeepSeek has demonstrated near-equivalent AI capabilities with far less compute. The author contends that the real existential threat comes from bad actors weaponizing AI, not from China as a nation-state, and that US-China cooperation on AI safety governance is essential.
Key Points
- •The capability gap between US and Chinese AI models has essentially closed, with Chinese models achieving near-equivalent results using a fraction of the compute resources.
- •US export controls on semiconductors have backfired by accelerating China's drive toward self-sufficiency, with even Commerce Secretary Raimondo calling it a 'fool's errand.'
- •Framing AI competition as a zero-sum national security race ignores the greater asymmetric threat from non-state bad actors and extremist groups weaponizing AI.
- •The real risk mirrors nuclear arms dynamics: nation-states are deterrable, but rogue actors willing to deploy AI for harm are far more dangerous.
- •The US and China should prioritize joint identification and mitigation of AI threats rather than purely competitive postures.
Cached Content Preview
HTTP 200Fetched Mar 20, 202620 KB
[Skip to Content](https://www.technologyreview.com/2025/01/21/1110269/there-can-be-no-winners-in-a-us-china-ai-arms-race/#content)
The United States and China are entangled in what many have dubbed an “ [AI arms race](https://www.forbes.com/sites/drewbernstein/2024/08/28/who-is-winning-the-ai-arms-race/).”
In the early days of this standoff, US policymakers drove an agenda centered on “winning” the race, mostly from an economic perspective. In recent months, leading AI labs such as [OpenAI](https://www.reuters.com/technology/artificial-intelligence/openai-urges-us-prioritize-ai-funding-regulation-stay-ahead-china-2025-01-13/) and [Anthropic](https://www.wsj.com/opinion/trump-can-keep-americas-ai-advantage-china-chips-data-eccdce91) got involved in pushing the narrative of “beating China” in what appeared to be an attempt to align themselves with the incoming Trump administration. The belief that the US _can_ win in such a race was based mostly on the early advantage it had over China in advanced GPU compute resources and the effectiveness of AI’s [scaling laws](https://openai.com/index/scaling-laws-for-neural-language-models/).
But now it appears that access to large quantities of advanced compute resources is no longer the defining or sustainable advantage many had thought it would be. In fact, 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](https://venturebeat.com/ai/deepseeks-first-reasoning-model-r1-lite-preview-turns-heads-beating-openai-o1-performance/) while using only a small fraction of the compute resources available to the leading Western labs.
The AI competition is increasingly being framed within narrow national security terms, as a zero-sum game, and influenced by assumptions that a future war between the US and China, centered on Taiwan, is inevitable. The US has employed [“chokepoint” tactics](https://www.cnn.com/2023/10/18/tech/us-china-chip-export-curbs-intl-hnk/index.html) to limit China’s access to key technologies like advanced semiconductors, and China has responded by accelerating its efforts toward self-sufficiency and indigenous innovation, which is causing US efforts to backfire.
Recently even outgoing US Secretary of Commerce Gina Raimondo, a staunch advocate for strict export controls, finally admitted that using such controls to [hold back China’s progress on AI and advanced semiconductors is a “fool’s errand.”](https://www.msn.com/en-us/politics/government/raimondo-says-holding-back-china-in-chips-race-is-a-fool-s-errand/ar-AA1witlV) Ironically, the unprecedented export control packages targeting China’s semiconductor and AI sectors have unfolded alongside tentative [bilateral and multilateral engagements](https://www.brookings.edu/articles/the-bletchley-park-process-could-be-a-building-block-for-global-cooperation-on-ai-safety/) to establish AI safety st
... (truncated, 20 KB total)Resource ID:
2fa332509e5f3ce0 | Stable ID: ZjhiMjQ1Y2