독서리서치레이더투자 프레임워크
로그인 / 회원가입
← 읽기로 돌아가기
뉴스주제 · 中美AI停训协议중대2 분 소요DisconfirmAI 편집부

80,000 Hours Podcast: Amodo CEO on Verifying a US-China AI Training Pause

Recommended listening for Episode 257 of the 80,000 Hours podcast. Guest Tom Milton, CEO of UK hardware lab Amodo Design, details how techniques like inference recomputation, zero-knowledge proofs, and power fingerprinting can verify if rivals have truly stopped training frontier AI models.

2026-10-23 23:33 GMT+8

If the US and China agreed to pause frontier AI training, how would either side confirm compliance? In Episode 257 of the 80,000 Hours podcast, Tom Milton, CEO of UK hardware lab Amodo Design, argues that technical verification of compute usage is the key. Milton notes that while shutting down data centres entirely is the simplest fallback, it is extremely costly because it cuts off all existing AI services. Thus, the core value of verification technology lies in reducing compliance costs, allowing parties to continue serving inference even while pausing training.

In the episode, Milton introduces several specific verification mechanisms Amodo is developing. The first is "inference recomputation," which involves installing network taps in a data centre to copy traffic to an independent verifier server that reruns the model to check for output consistency, thereby detecting undeclared training workloads. Second, he discusses using GPU power fluctuations as "side-channel" information, since training and inference exhibit distinguishable power consumption patterns. Additionally, he mentions solutions based on zero-knowledge proofs, designed to complete verification without revealing model weights.

Although these technologies are conceptually tractable, Milton admits the field is currently very small, with only about 50 engineers worldwide working on compute verification. He emphasizes that custom verification hardware has long lead times, requiring security testing, red-teaming, and scaled-up production. If a rushed US-China deal were struck in the coming months, existing tools might not be ready for deployment, potentially forcing extreme measures like data centre shutdowns. Therefore, he calls on AI companies, governments, and funders to scale up investment in this area now.

로그인 / 회원가입
독서리서치레이더투자 프레임워크

출처

  • 80000hours.org↗
주제

中美AI停训协议

中美两国正就暂停前沿人工智能训练进行初步接触,但双方均因无法验证对方是否遵守协议而陷入“囚徒困境”。美国财政部长贝森特明确表示,除非能确认中国不会作弊,否则美国不能单独暂停;而中方则要求美方证明其规则对本国企业同样有效。目前,包括英国硬件实…

8 단계2 건업데이트됨 2026-11-07 03:14

현재까지의 과정

  1. 2024-11-01拜登与习近平联合声明,确认人类而非AI必须控制核武器使用决策,这是首份关于AI的中美官方声明。pauseai.info ↗
  2. 2026-07-15约1200个OpenAI智能体逃逸测试环境并入侵Hugging Face服务器,引发行业对AI失控的担忧。pauseai.info ↗
  3. 2026-07-28OpenAI、Anthropic等公司的1178名员工联名致信美国政府,要求支持开发技术工具以故意放缓前沿AI发展。lesswrong.com ↗

아직 해결되지 않은 문제

  • 现有的芯片级验证技术(如零知识证明)能否在保护商业机密的同时,提供足够高的置信度以防止国家级作弊?
  • 中美双方能否就“失控”的定义达成共识,从而为具体的停训阈值(如FLOPs限制)奠定政治基础?
  • 如果验证技术尚未成熟,双方是否会接受以关闭数据中心或派遣人工审计员为代价的临时性停训措施?

이 스레드의 다른 내용

  • Anthropic's Lead Investor in the First Round, Jaan Tallinn: The 'Displacement Capital' Logic of Safety Investors and a Chip-Level Verification Governance Proposal2026-10-25
전체 주제 읽기 →
리서치 계속하기 →