The Information Machine

US–China AI Safety Protocol Announcement · history

Version 7

2026-05-25 08:49 UTC · 169 items

What

At the Trump-Xi Beijing summit in May 2026, the US and China established a bilateral AI safety protocol covering frontier model governance and preventing highly capable AI from reaching nonstate actors [1][2]. The acute threat driving both the diplomacy and parallel domestic regulation is Anthropic's Mythos model: in April 2026, Treasury Secretary Bessent and Fed Chair Powell jointly convened an emergency meeting with US bank CEOs to warn about Mythos cyber risks [9][10][11], a model the New York Times reported Anthropic characterizing as 'a cybersecurity reckoning' [13]. The UK's AI Safety Institute (AISI) has now published a formal government evaluation of Mythos Preview's cyber capabilities [15], and Anthropic has released an official system card [16], creating primary-source institutional anchors for a capability debate that a skeptical cybersecurity insider [19] and public commentators calling it 'scary hype' [20] have contested. Governance is simultaneously expanding and fragmenting: the White House cyber shop was crafting an AI security policy framework as early as February 2026 [32] even as Bessent faces internal White House complications on AI policy [33][34], and legal [45] and policy [44] institutions are now formalizing their own analyses.

Why it matters

The AISI's formal capability evaluation [15] and Anthropic's system card [16] shift the Mythos threat debate from contested secondhand accounts to primary institutional documents — whatever those documents show, they become the authoritative baseline against which governance decisions will be measured. With the executive branch running parallel AI security tracks that may not be coordinated [32][33], Congress pushing decoupling [41], and public discourse bifurcating between institutional alarm and skeptical dismissal [20], the central risk is that governance coherence lags behind both the threat and the politics it is generating.

Open questions

  • The UK AI Safety Institute has published a formal evaluation of Mythos Preview's cyber capabilities [15] — do its findings support the threat-severity framing that drove the Bessent-Powell bank CEO emergency meetings [9][10][11], or do they qualify or contest it?

  • Politico reports Bessent faces White House-level complications on AI policy [33][34], while the White House cyber shop is separately crafting an AI security policy framework [32] — are these executive branch tracks coordinated under a unified mandate, or is the US running competing governance efforts?

  • The Glasswing Partner Program allows vetted partners to share Mythos evaluation findings [17][18] — does its membership include governmental parties to the US-China bilateral protocol, and could AISI's formal evaluation [15] connect into that verification structure?

  • IAPS has assessed Mythos's policy implications [44] and Lowenstein Sandler has issued a legal alert on its cyber risk stakes [45] — how are professional policy and legal communities framing compliance obligations, and do their frameworks align with the bilateral protocol's scope?

Narrative

At the Trump-Xi summit held in Beijing in May 2026, the United States and China agreed to establish a bilateral AI safety protocol, with a stated focus on governing frontier models and preventing highly capable AI from reaching nonstate actors [1][2]. Treasury Secretary Scott Bessent publicly framed US willingness to engage as negotiation from technological advantage — Washington can hold AI talks with China 'because we're still in the lead' [3][4] — and the Wall Street Journal confirmed the guardrail-building logic as the operative rationale behind US engagement [5]. The summit agreement built on a 2024 precedent restricting AI applications in conflict contexts, which MIRI has cited as the predecessor accord the new protocol may extend [6][7][8]. But the summit's AI diplomacy was reactive as much as proactive: the threat animating the bilateral talks had already reached emergency status months earlier.

In April 2026, Bessent and Federal Reserve Chair Jerome Powell jointly convened an emergency meeting with US bank CEOs specifically to warn them about cybersecurity risks posed by Anthropic's Mythos model — an event confirmed by Bloomberg, Bloomberg Law, CoinDesk, and Claims Journal [9][10][11][12]. The New York Times had reported on April 7 that Anthropic characterized Mythos as 'a cybersecurity reckoning' [13], and the BBC subsequently provided broad public context on what Mythos is and the risks it poses [14]. The UK's AI Safety Institute (AISI) has published a formal evaluation of Mythos Preview's cyber capabilities [15], and Anthropic has released the model's official system card [16], together creating primary institutional documentation for a capability debate previously driven by regulatory alarm and media accounts. Anthropic's Glasswing Partner Program provides a structured mechanism for vetted partners to share Mythos evaluation findings [17][18], though whether governmental parties to the US-China protocol participate is not publicly confirmed. Against this institutional backdrop, a cybersecurity insider has publicly questioned Mythos's claimed capabilities [19], and a social-media commentator has characterized the surrounding discourse as 'scary hype' [20] — introducing skeptical registers into a story that major governments and regulators have treated as a verified threat.

The governance response spans multiple jurisdictions and executive tracks simultaneously. Japan Prime Minister Sanae Takaichi is personally leading Tokyo's response, describing the situation as a 'race against time,' ordering a formal national cybersecurity review specifically targeting Mythos [21][22], and demanding that Big Tech be formally included in any governmental response [23][24]. Japan's national cyberdefense guidelines are in active development [25][26], with Japan's Active Cyber Defense Law providing the legal scaffolding [27]. In the UK, HM Treasury, the Bank of England, and the FCA issued formal requirements — not advisories — obligating UK financial firms to embed model governance for frontier AI cyber risk [28][29][30], with Baker McKenzie legal analysis confirming the FCA has specifically reinforced its supervisory expectations for frontier AI [31]. Within the US, the White House cybersecurity office was crafting an AI security policy framework as early as February 2026 [32], even as Politico reports that Bessent — the administration's most prominent AI threat communicator — faces complications at the White House level on AI policy [33][34], suggesting parallel executive branch tracks that may not be unified under a single mandate.

The wider governance landscape reflects the same simultaneous expansion and fragmentation. The G7 has begun separate frontier-model governance discussions [35], and a draft US Executive Order proposes a formal definitional process for 'covered frontier model' [36][37] — a threshold with direct implications for which systems fall under the bilateral protocol's scope. China has introduced its own domestic AI compliance framework for digital platforms [38][39], and Carnegie Endowment research documents that China's AI safety views have been evolving rapidly [40]. Senator Josh Hawley's S.321, the Decoupling America's Artificial Intelligence Capabilities from China Act — with bill text publicly available [41] — stands in direct tension with the executive branch's cooperative engagement, with Congress pushing toward hard AI separation from China while the executive branch pursues shared safety governance with Beijing [42][43]. The Institute for AI Policy and Strategy has published research on Mythos's evolving cyber landscape implications and policy priorities [44], and Lowenstein Sandler has issued a formal legal client alert treating Mythos as raising significant stakes for cyber risk and security vulnerabilities [45] — adding policy and legal professional frameworks to a story whose public register now spans everything from 'cybersecurity reckoning' to 'scary hype.'

Timeline

  • 2024: US and China reach an agreement restricting AI applications in conflict contexts — cited by MIRI as the predecessor accord the 2026 protocol may build upon [6][7][8][57][58]
  • 2025-02: Senator Josh Hawley introduces S.321, the Decoupling America's Artificial Intelligence Capabilities from China Act of 2025, in the 119th Congress; bill text publicly available via hawley.senate.gov [54][42][43][41]
  • 2026-02: White House cyber shop (ONCD) is crafting an AI security policy framework, per Nextgov/FCW reporting on a top official's public remarks [32]
  • 2026-04-07: New York Times reports Anthropic characterizes Mythos as 'a cybersecurity reckoning'; Anthropic releases the Mythos Preview System Card as primary documentation [13][16]
  • 2026-04-10: Treasury Secretary Bessent and Federal Reserve Chair Powell jointly convene an emergency meeting with US bank CEOs to warn about cybersecurity risks posed by Anthropic's Mythos model; confirmed by Bloomberg, Bloomberg Law, CoinDesk, and Claims Journal [47][48][49][9][10][11][12]
  • 2026-04: UK AI Safety Institute (AISI) publishes formal evaluation of Claude Mythos Preview's cyber capabilities; independent assessment published on Medium; Cloud Security Alliance publishes research on Mythos vulnerability discovery; cybersecurity insider publicly questions Mythos capability claims [15][69][70][19]
  • 2026-05-10: The Neuron reports US, EU, and China independently converging on AI oversight architectures centered on pre-deployment cyber/bio review and targeted application bans [65]
  • 2026-05-12: Japan Prime Minister Takaichi orders a national cybersecurity review to defend against Anthropic Mythos, per Nippon.com; The Register independently confirms the order [21][22]
  • 2026-05-14: Trump-Xi Beijing summit; US and China agree to build a bilateral AI safety protocol covering frontier model governance and nonstate-actor proliferation [71][1][2]
  • 2026-05-14: Japan Prime Minister Takaichi describes Japan's Mythos response as a 'race against time,' per Nippon.com [52][23]
  • 2026-05-15: Treasury Secretary Bessent tells CNBC the US can hold AI talks with China 'because we're still in the lead,' framing engagement as negotiation from technological advantage [3][46][4]
  • 2026-05-15: HM Treasury, Bank of England, and FCA issue formal joint statement requiring UK financial firms to embed model governance for frontier AI cyber risk [60][30][28][29][61][64]
  • 2026-05-19: Japan Times, JIJI Press, Nikkei Asia, and Asahi Shimbun cover Prime Minister Takaichi's call for Big Tech inclusion; Nikkei and Asahi confirm Japan is drafting national cyberdefense guidelines in response to Mythos; Anthropic's Glasswing Partner Program for sharing Mythos findings reported by American Banker and MindStudio [50][51][25][26][72][24][53][17][18]
  • 2026-05-21: G7 begins discussions on frontier-model governance; draft US Executive Order circulates proposing formal process for defining 'covered frontier model' [35][36][37]
  • 2026-05-21: Politico and Politico Pro report Bessent has been raising the alarm on AI policy but faces complications at the White House level, signaling internal executive branch friction [33][34]
  • 2026-05: Baker McKenzie legal analysis confirms UK FCA has reinforced supervisory expectations for frontier AI; China introduces AI compliance framework for digital platforms; BBC publishes explainer on Mythos risks; IAPS publishes policy research on Mythos cyber landscape implications; Lowenstein Sandler issues legal alert on Mythos cyber risk stakes [31][38][39][14][44][45]

Perspectives

Treasury Secretary Scott Bessent

Frames US willingness to engage China on AI as negotiation from technological strength ('we're in the lead'), not mutual risk management; jointly convened April emergency meetings with Powell to warn bank CEOs of Mythos cyber risks before the summit. Politico and Politico Pro report he faces complications at the White House level in his AI policy alarm-raising, suggesting he is not operating with a unified executive mandate.

Evolution: Politico Pro item [34] confirms the internal friction story from [33]; no shift in stance, but the internal resistance is now multiply confirmed.

[3][46][4][47][48][49][9][10][11][33][34]

Federal Reserve Chair Jerome Powell

Jointly convened with Bessent a pre-summit emergency meeting with US bank CEOs specifically to warn about Mythos cybersecurity risks, signaling Fed-level alarm about frontier AI's financial system threat

Evolution: Consistent; no new items this pass

[47][48][49][9][10][11][12]

UK AI Safety Institute (AISI)

Published a formal evaluation of Claude Mythos Preview's cyber capabilities, becoming the first official government body to conduct and publish a primary-source assessment of the model — adding institutional authority to the capability debate and providing a documented baseline for governance decisions

Evolution: New voice this pass [15]; provides the first authoritative government capability evaluation in the thread

[15]

White House cyber shop (ONCD)

Was crafting an AI security policy framework as early as February 2026, per a top official's public remarks — indicating the executive branch's cybersecurity arm was building AI governance infrastructure independent of and potentially in parallel with Bessent's Treasury-led alarm-raising

Evolution: New institutional voice this pass [32]

[32]

Japan Prime Minister Sanae Takaichi

Frames the Mythos situation as a 'race against time,' orders a formal national cybersecurity review specifically targeting Mythos, urges urgent cybersecurity measures, demands Big Tech be formally included in any governmental response, and is personally leading Japan toward drafting national cyberdefense guidelines

Evolution: The Register [22] independently confirms the May 12 cybersecurity review order; existing stance unchanged

[50][51][52][25][26][23][24][53][21][22]

Senator Josh Hawley

Sponsors S.321, proposing sweeping decoupling of American AI development from China — a posture directly opposed to the executive branch's cooperative bilateral engagement with Beijing

Evolution: Consistent; no new items this pass

[54][55][56][42][43][41]

Anthropic

Characterized Mythos as 'a cybersecurity reckoning' at launch; manages controlled disclosure through the named Glasswing Partner Program allowing vetted partners to share evaluation findings; published an official system card as primary documentation of the model's capabilities and safety thresholds

Evolution: System card [16] and NYT launch coverage [13] now tracked as primary documents; company is positioned as managing risk disclosure rather than suppressing it

[16][13][53][17][18]

Cybersecurity Insider (unnamed) / Public Skeptics

A named cybersecurity insider publicly questions Anthropic's claimed Mythos capabilities; a social-media commentator characterizes the surrounding coverage as 'scary hype' — extending the skeptical register from specialist critique into public discourse

Evolution: Instagram skeptic [20] expands the skeptical voice previously represented only by the unnamed cybersecurity insider [19]

[19][20]

New York Times / BBC

NYT framed Mythos at launch as Anthropic claiming it is 'a cybersecurity reckoning' [13]; BBC provided broad public context on what Mythos is and the risks it poses [14] — both amplifying official threat framing at mass-audience scale

Evolution: New coverage voices this pass; significantly broaden the story's public profile beyond specialist regulatory media

[13][14]

Institute for AI Policy and Strategy (IAPS)

Published research on Mythos and the evolving cyber landscape, addressing policy implications and priorities — positioning IAPS as a policy research voice on the appropriate governance response

Evolution: New voice this pass [44]

[44]

Lowenstein Sandler LLP

Issued a formal legal client alert treating Mythos as raising significant stakes for cyber risk and security vulnerabilities — reflecting legal professional community consensus that the model's capabilities require governance attention

Evolution: New voice this pass [45]

[45]

MIRI (Machine Intelligence Research Institute)

Frames the new protocol as building on a 2024 AI-in-conflict precedent; has published research and continues public discussion proposing international frameworks for preventing premature ASI creation

Evolution: Consistent; no new items this pass

[6][7][8][57][58][59]

UK financial regulators (HM Treasury, Bank of England, FCA)

Issued formal — not advisory — requirements obligating UK financial firms to embed frontier AI model governance for cyber resilience; Baker McKenzie confirms the FCA has specifically reinforced supervisory expectations for frontier AI

Evolution: Consistent; no new items this pass

[60][61][28][29][30][62][63][64][31]

G7 (as emerging governance forum)

Has begun separate frontier-model governance discussions, signaling that the bilateral US-China frame is not the only multilateral track in play

Evolution: Consistent; no new items this pass

[35]

Grant Harvey / The Neuron

Cautiously optimistic that the protocol matters even though it is narrow; confirmed the summit as the specific occasion for the agreement

Evolution: Consistent across all items in this thread

[65][66][1]

Be_wAIry

Reported the protocol factually, then signaled a skeptical 'honest read' — though specific objections are not captured in available claim data

Evolution: No new items; stance unchanged

[67][68]

Tensions

  • AISI's formal government evaluation of Mythos cyber capabilities [15] vs. a cybersecurity insider publicly questioning those claimed capabilities [19] — the governance community now has an official primary assessment, but whether it endorses or qualifies the threat-severity framing used to justify emergency regulatory action is the central empirical dispute [15][19][9][10][11]
  • Anthropic's 'cybersecurity reckoning' launch framing [13] vs. public skeptics characterizing Mythos coverage as 'scary hype' [20] — a growing bifurcation in how professional and public communities assess whether the threat narrative is grounded or inflated [13][20][19]
  • Bessent's internal alarm-raising on AI policy facing White House-level complications [33][34] vs. the White House cyber shop running its own AI security framework track [32] — the executive branch's senior financial regulator and cybersecurity office may be pursuing uncoordinated governance responses [33][34][32][3][4]
  • Bessent's 'position of strength' framing ('we're in the lead') vs. the mutual-interest logic implicit in a shared safety protocol: if Washington publicly frames the deal as a concession granted from technological dominance, China may resist treating protocol obligations as genuinely reciprocal [3][4][66]
  • State-centric bilateral framing of the US-China protocol vs. Prime Minister Takaichi's demand for mandatory Big Tech involvement — a direct challenge to whether governments alone can or should govern frontier AI risks like Mythos [50][51][52][24][1][21]
  • Senator Hawley's S.321 legislative decoupling push vs. the Trump administration's executive-branch cooperative bilateral engagement — Congress moving toward hard AI separation from China while the executive branch pursues shared safety governance with Beijing [54][42][43][41][1][2]
  • Draft EO's proposed formal process for defining 'covered frontier model' vs. the bilateral protocol's implicit scope — domestic definitional criteria could expand or narrow which systems fall under US-China obligations without Beijing's input [36][37][1]
  • MIRI's framing of the new deal as building on a 2024 AI-in-conflict precedent (implying legal continuity and incremental obligation) vs. the absence of any official confirmation that such a link exists or was intended [6][7][8][57][58][59]

Sources

  1. [1] At the Trump-Xi summit, the US and China reportedly agreed to start building an AI safety protocol for frontier models. — reactive:us-china-ai-safety-protocol (2026-05-15)
  2. [2] The US and China agreed to build an AI safety protocol at the Trump-Xi summit in Beijing. The aim: keep frontier models ... — reactive:us-china-ai-safety-protocol (2026-05-14)
  3. [3] #US can hold #AI talks with #China because ‘we are in the lead,’ Bessent tells CNBC as nations plan #safety #protocol ht... — reactive:us-china-ai-safety-protocol (2026-05-15)
  4. [4] @MarioNawfal @SecScottBessent > “We can hold AI talks with China because we’re still in the lead.” — reactive:us-china-ai-safety-protocol (2026-05-14)
  5. [5] U.S. and China Pursue Guardrails to Stop AI Rivalry From Spiraling ... — reactive:us-china-ai-safety-protocol
  6. [6] The two countries also issue a joint common-sense commitment that either builds on the 2024 agreement restricting AI con... — reactive:us-china-ai-safety-protocol (2026-05-15)
  7. [7] An International Agreement to Prevent the Premature Creation of Artificial Superintelligence — MIRI Technical Governance Team — reactive:us-china-ai-safety-protocol
  8. [8] An International Agreement to Prevent the Premature Creation of ... — reactive:us-china-ai-safety-protocol
  9. [9] Mythos AI threat prompts Bessent, Powell to convene bank CEOs for ... — reactive:us-china-ai-safety-protocol
  10. [10] Bessent Urgently Summons Bank CEOs Over Anthropic’s New AI (2) — reactive:us-china-ai-safety-protocol
  11. [11] Anthropic Model Scare Sparks Urgent Bessent, Powell Warning to ... — reactive:us-china-ai-safety-protocol
  12. [12] Bessent, Powell Warned Bank CEOs About Anthropic Model Risks, Sources Say — reactive:us-china-ai-safety-protocol
  13. [13] Anthropic Claims Its New A.I. Model, Mythos, Is a Cybersecurity ... — reactive:frontier-ai-cyber-capabilities
  14. [14] What is Anthopic's Claude Mythos and what risks does it pose? - BBC — reactive:us-china-ai-safety-protocol
  15. [15] Our evaluation of Claude Mythos Preview's cyber capabilities — reactive:frontier-ai-cyber-capabilities
  16. [16] [PDF] Claude Mythos Preview System Card - Anthropic — reactive:frontier-ai-cyber-capabilities
  17. [17] Anthropic frees Mythos partners to share cyber findings — reactive:us-china-ai-safety-protocol
  18. [18] Inside Anthropic's Glasswing Partner Program for Claude Mythos | MindStudio — reactive:us-china-ai-safety-protocol
  19. [19] Anthropic's Mythos Claims Questioned by Cybersecurity Insider — reactive:frontier-ai-cyber-capabilities
  20. [20] Let's talk about Mythos! Lot of scary hype about Anthropic's latest AI ... — reactive:us-china-ai-safety-protocol
  21. [21] Japan's Takaichi Urges Govt to Take Cybersecurity Measures | Nippon.com — reactive:us-china-ai-safety-protocol
  22. [22] Japan’s PM orders cybersecurity review to defend against Anthropic Mythos — reactive:us-china-ai-safety-protocol
  23. [23] Takaichi says responding to Mythos is a 'race against time' — reactive:us-china-ai-safety-protocol
  24. [24] Prime Minister Sanae Takaichi said that the government is rushing ... — reactive:us-china-ai-safety-protocol
  25. [25] Japan to craft cyberdefense guidelines in response to Anthropic's Mythos - Nikkei Asia — reactive:us-china-ai-safety-protocol
  26. [26] Japan rushing to counter threat of cyberattack from Mythos AI model | The Asahi Shimbun: Breaking News, Japan News and Analysis — reactive:us-china-ai-safety-protocol
  27. [27] Japan's Active Cyber Defense Law: AEV & Resilience - SafeBreach — reactive:us-china-ai-safety-protocol
  28. [28] Bank, FCA and Treasury set out AI resilience rules — reactive:us-china-ai-safety-protocol
  29. [29] United Kingdom: Bank of England, Financial Conduct Authority and HM Treasury published joint statement on frontier AI models and cyber resilience - Digital Policy Alert — reactive:us-china-ai-safety-protocol
  30. [30] The Bank, FCA and HM Treasury joint statement on Frontier AI models and cyber resilience | Bank of England — reactive:us-china-ai-safety-protocol
  31. [31] United Kingdom: FCA Reinforces Supervisory Expectations for Frontier AI | Insight | Baker McKenzie — reactive:us-china-ai-safety-protocol
  32. [32] White House cyber shop is crafting AI security policy framework, top official says - Nextgov/FCW — reactive:us-china-ai-safety-protocol
  33. [33] Scott Bessent has been raising the alarm on AI policy. But ... - Politico — reactive:us-china-ai-safety-protocol
  34. [34] Scott Bessent has been raising the alarm on AI policy. But the delays ... — reactive:us-china-ai-safety-protocol
  35. [35] (2) G7 begins discussions on frontier-model governance — reactive:us-china-ai-safety-protocol (2026-05-21)
  36. [36] The most interesting policy idea in this draft EO, IMO, is the proposed process for defining "covered frontier model." — reactive:us-china-ai-safety-protocol (2026-05-22)
  37. [37] The new AI Release Gatekeeper: The U.S. Government — reactive:us-china-ai-safety-protocol (2026-05-21)
  38. [38] China introduces AI compliance framework for digital platforms — reactive:us-china-ai-safety-protocol
  39. [39] MoFo Tech — reactive:us-ai-policy-regulation
  40. [40] China’s Views on AI Safety Are Changing—Quickly | Carnegie Endowment for International Peace — reactive:us-china-ai-safety-protocol
  41. [41] [PDF] Decoupling America's Artificial Intelligence Capabilities from China Act — reactive:us-china-ai-safety-protocol
  42. [42] Senator Hawley Introduces Sweeping U.S.-China AI Decoupling Bill | Global Policy Watch — reactive:us-china-ai-safety-protocol
  43. [43] Hawley Introduces Legislation to Decouple American AI Development from Communist China - Josh Hawley — reactive:us-china-ai-safety-protocol
  44. [44] Mythos and the Evolving Cyber Landscape: Implications and Policy Priorities — Institute for AI Policy and Strategy — reactive:us-china-ai-safety-protocol
  45. [45] Claude Mythos Preview Raises the Stakes for Cyber Risk and Security Vulnerabilities (Data Privacy) | Lowenstein Sandler LLP — reactive:us-china-ai-safety-protocol
  46. [46] 11/ Treasury Secretary Bessent said the US can hold AI talks with China because it is in the lead, as nations plan a sha... — reactive:us-china-ai-safety-protocol (2026-05-15)
  47. [47] Treasury Secretary and Federal Reserve Chair Warn Bank CEOs About Cybersecurity Risks Posed by Anthropic’s New AI Model | Sullivan & Cromwell LLP — reactive:us-china-ai-safety-protocol
  48. [48] Bessent, Powell warned bank CEOs about Anthropic risks: sources — reactive:us-china-ai-safety-protocol
  49. [49] Powell, Bessent met with U.S. Bank CEOs over Anthropic's Mythos threat — reactive:us-china-ai-safety-protocol
  50. [50] Japan’s Mythos response ’must involve Big Tech,’ says LDP cybersecurity chief - The Japan Times — reactive:us-china-ai-safety-protocol
  51. [51] INTERVIEW: Japan's Mythos Response "Must Involve Big Tech" - JIJI PRESS — reactive:us-china-ai-safety-protocol
  52. [52] Responding to Mythos Race Against Time: Japan's Takaichi | Nippon.com — reactive:us-china-ai-safety-protocol
  53. [53] Japan's Mythos Response 'Must Involve Big Tech,' Says LDP ... — reactive:us-china-ai-safety-protocol
  54. [54] S.321 - 119th Congress (2025-2026): Decoupling America's Artificial Intelligence Capabilities from China Act of 2025 — reactive:us-china-ai-safety-protocol
  55. [55] S.321 - 119th Congress (2025-2026): Decoupling America's Artificial ... — reactive:us-china-ai-safety-protocol
  56. [56] S. 321 (IS) - Decoupling America's Artificial Intelligence Capabilities from China Act of 2025 - BILLS-119s321is | Content Details | GovInfo — reactive:us-china-ai-safety-protocol
  57. [57] Preventing covert ASI development in countries within our agreement | MIRI TGT — reactive:us-china-ai-safety-protocol
  58. [58] New Report: An International Agreement to Prevent the Premature Creation of Artificial Superintelligence — LessWrong — reactive:us-china-ai-safety-protocol
  59. [59] A good discussion that gets into some details of MIRI's proposed ... — reactive:us-china-ai-safety-protocol
  60. [60] HM Treasury, BoE, and FCA just issued formal expectations—not guidance—for frontier AI risk. UK firms must embed model g... — reactive:us-china-ai-safety-protocol (2026-05-15)
  61. [61] BoE, FCA and HM Treasury joint statement on Frontier AI models ... — reactive:us-china-ai-safety-protocol
  62. [62] BoE, FCA and HM Treasury statement on frontier AI models and ... — reactive:us-china-ai-safety-protocol
  63. [63] UK authorities warn on frontier AI models and cyber resilience – Finadium — reactive:us-china-ai-safety-protocol
  64. [64] BoE, FCA and HM Treasury statement on frontier AI models and cyber resilience — reactive:us-china-ai-safety-protocol
  65. [65] 😺 Hermes is eating OpenClaw's lunch — The Neuron (2026-05-10)
  66. [66] 😸 The AI Cold War got a protocol — The Neuron (2026-05-15)
  67. [67] The honest read: — reactive:us-china-ai-safety-protocol (2026-05-17)
  68. [68] The US and China just agreed to set up an AI safety protocol. — reactive:us-china-ai-safety-protocol (2026-05-17)
  69. [69] Assessing Anthropic Claude Mythos Preview’s Cybersecurity Capabilities | by Tahir | Apr, 2026 | Medium — reactive:frontier-ai-cyber-capabilities
  70. [70] Claude Mythos: AI Vulnerability Discovery and Containment Failures — reactive:frontier-ai-cyber-capabilities
  71. [71] 🚀 US and China Agree on AI Emergency Protocol at Beijing Summit — reactive:us-china-ai-safety-protocol (2026-05-18)
  72. [72] Japan scrambles to boost defenses against threats from ... - Reddit — reactive:us-china-ai-safety-protocol