Version 24

What

Why it matters

Open questions

• Does Lutnick's lift letter formally rescind the BIS export licensing requirement and the Pentagon's supply-chain designation, or does it address only the access restriction while leaving the formal designations in place [14][5][19]?

• What happens to Legion LegalTech's federal lawsuit now that the controls are lifted — does the case proceed on a record of constitutional harm, or does restoration moot the action [18][17]?

• Zvi reports CAISI was under a stop-work order throughout the suspension [17] — does the White House now have in-house AI technical capacity to evaluate future model-risk claims, or will ad hoc reliance on industry briefings continue?

• OpenAI secured a Pentagon contract within hours of the Anthropic ban [10][11] — has the DoD's supply-chain designation been formally withdrawn, and does OpenAI's deal remain in place now that Anthropic's models are restored?

Narrative

Timeline

• 2026-04-22: Alibaba and Alibaba Qwen begin a model-cloning campaign against Claude, ultimately generating 28.8 million exchanges via approximately 25,000 fraudulent accounts through June 5. [12]
• 2026-06-10: Anthropic sends a confidential letter to Senators Tim Scott and Elizabeth Warren disclosing the Alibaba cloning campaign, the day before a Senate AI hearing. [12]
• 2026-06-12: Amazon researchers demonstrate a jailbreak on Fable 5; Jassy personally briefs senior Trump administration officials in conversations that directly preceded the government action. [2][3][28]
• 2026-06-12: Commerce Department issues export control directive for all foreign nationals; Lutnick tells Amodei 'That's the point' when informed both models would have to go offline. [20][29][1][4]
• 2026-06-13: Anthropic disables Fable 5 and Mythos 5 globally after determining nationality-based restriction was not achievable on short notice. [1]
• 2026-06-16: Moussouris assesses the triggering behavior as 'the model working as intended' for cyberdefense; Luta Security publishes a formal argument that the controls harm US cyber defense. [6][22]
• 2026-06-17: Lutnick sends a formal BIS letter to Amodei requiring an export license before international deployment. [5][4]
• 2026-06-20: Trump states he no longer views Anthropic as a national security threat, with no formal policy change announced. [9][30]
• 2026-06-21: Pentagon labels Anthropic a 'supply chain risk,' adding a DoD procurement designation alongside the BIS export licensing controls. [19]
• 2026-06-24: Legion LegalTech files a federal lawsuit against the US government — the first formal judicial challenge to the directive. [18][25][26]
• 2026-06-24: Zvi reports the NSA lost its own Mythos access; NSA red-team results cited as justification involved authorized insider access on air-gapped systems, not external attack. [8]
• 2026-06-25: Anthropic publicly calls for punishment of Alibaba for the 28.8-million-exchange cloning campaign. [12]
• 2026-06-26: Trump orders military contractors and federal agencies to cease business with Anthropic; OpenAI secures a Pentagon deal within hours. [21][11]
• 2026-06-27: Lutnick authorizes Mythos 5 for more than 100 approved institutions; Anthropic confirms Mythos redeployment to US critical infrastructure organizations. [16][31]
• 2026-06-29: Zvi Mowshowitz calls a WSJ report claiming China's GLM-5.2 matched Mythos 'obvious nonsense,' arguing Mythos's autonomous large-scale vulnerability identification without being pointed at specific targets remains unmatched. [24]
• 2026-06-30: Commerce Department lifts export controls on both Fable 5 and Mythos 5; Lutnick sends a formal letter confirming the lift; Anthropic announces access restoration begins July 1. [13][14][15]

Perspectives

Tensions

• Anthropic, Moussouris, and Luta Security assess the triggering behavior as standard defensive security work on engineered fake code; the government treated it as justifying full model suspension and BIS licensing, and never publicly addressed the expert counter-assessment — even when lifting the controls. [6][22][4][20][5][14]
• Trump publicly stated he no longer views Anthropic as a national security threat, yet ordered military contractors and agencies to stop using Anthropic days later; the June 30 lift partially closes the gap but no administration statement reconciles the sequence. [9][21][11][14]
• The stated reinstatement path required fixing the jailbreak; Zvi argues this is technically impossible and the actual resolution was political accommodation, with no technical bar ever verified or announced. [4][7][27][17]
• Anthropic contests the export controls as technically disproportionate while simultaneously providing evidence that Alibaba extracted Claude's capabilities at scale via 28.8 million fraudulent exchanges — evidence that supports the government's broader China-threat framing even if not the specific triggering incident. [12][20][6][5]
• The Wall Street Journal reported China's GLM-5.2 matched Mythos in cybersecurity; Zvi argues GLM-5.2 only replicates Mythos in narrow directed scenarios and Mythos's autonomous large-scale vulnerability identification remains unmatched. [24]