Version 13

What

Why it matters

Open questions

• What specific capabilities of Fable 5 or Mythos 5 triggered the June 12 directive, and does SK Telecom's Project Glasswing involvement explain the timing or only the jurisdiction? [18][16]

• Will Anthropic be permitted to re-enable Fable 5 and Mythos 5 after satisfying export control requirements, or does the suspension reflect a capability threshold the government considers permanently off-limits for global access? [17][19]

• Anthropic has not responded to Jeremy Howard's argument that its continued use of frontier models for its own research internally negates its slowdown call [11] — does the Fable 5/Mythos 5 suspension change that calculus?

• If Altman sees a major RSI breakthrough as reason to stay private [13][14], does the forced Anthropic model suspension change OpenAI's IPO or governance stance?

Narrative

Timeline

• 2026-05: Claude authors 80%+ of Anthropic's production code; per-engineer output reaches 8x the 2024 baseline; Claude Mythos Preview accelerates model-training code ~52x; Claude's coding success rate reaches 76%, up 50 points in six months. [2][34][1]
• 2026-06-03: OpenAI publishes 'Democratic Governance of Frontier AI,' acknowledging early RSI signs and proposing CAISI as a federal oversight body with mandatory evaluation authority. [33][3][30]
• 2026-06-04: Anthropic publishes 'When AI Builds Itself,' disclosing Claude's 80%+ code authorship and calling for a global coordinated slowdown in frontier AI development. [27][35][36]
• 2026-06-05: Critics note timing of Anthropic's slowdown call relative to its confidential S-1 filing at roughly $965 billion valuation. [9][35][10]
• 2026-06-08: Sam Altman predicts AI will conduct a significant fraction of OpenAI's research by March 2028; OpenAI confidentially files for an IPO. [24][25]
• 2026-06-09: All three major labs publicly endorse international coordination to slow frontier AI development; NSA reported using Claude Mythos for offensive cyber with approximately six embedded Anthropic engineers. [5][7]
• 2026-06-10: Geoffrey Irving founds Sequent; Jeremy Howard argues Anthropic's use of its own frontier models for research negates its slowdown call; Altman tells staff a major RSI breakthrough would favor staying private over pursuing the IPO. [23][11][13][14][26]
• 2026-06-11: US government directs CAISI to stop publishing public AI model evaluations; Amodei tells Bloomberg AI progress is now in the sharp-acceleration phase of an exponential curve. [6][28]
• 2026-06-12: Google DeepMind identifies four distinct technical pathways through which AGI could transition to ASI. [4]
• 2026-06-12: US government issues export control directive requiring Anthropic to block global access to Fable 5 and Mythos 5; SK Telecom's inclusion in Project Glasswing identified as the trigger. [37][16][18]
• 2026-06-13: Anthropic disables Fable 5 and Mythos 5 globally for all customers. [15][16][17]
• 2026-06-15: Jack Clark's Import AI 461 declares 'alignment is not on track'; Sequent confirms plans to raise $100-150M and scale to 40-80 employees. [22]
• 2026-06-17: Tech Policy Press asks whether the US has set an AI export control precedent by blocking Mythos from global access. [19]
• 2026-06-19: Dario Amodei discloses that testers of Anthropic's most powerful unreleased model recommended against release, describing it as a 'super weapon'; the characterization is clarified as a metaphor from early testers, not a literal licensing requirement. [20][21]

Perspectives

Tensions

• Howard argues Anthropic's continued use of its top-ranked model for frontier research negates its slowdown call; Anthropic has not responded. [11]
• Altman publicly endorses coordinated slowdown mechanisms while telling staff that a major RSI breakthrough would favor staying private — the two positions point in different directions on what OpenAI actually wants from RSI timing. [12][13][14]
• OpenAI's blueprint proposed CAISI as a federal body with public evaluation authority; the US government directed CAISI to stop publishing public AI model evaluations, removing the transparency mechanism the proposal depended on. [33][6]
• The US simultaneously embeds Anthropic engineers at the NSA for offensive cyber and issues export control directives that force Anthropic's most capable models offline globally — an incoherent regulatory posture toward the same company. [7][16][18]
• Anthropic frames its slowdown call as a genuine safety response; critics argue it is regulatory strategy timed to a confidential S-1 filing at roughly $965 billion, designed to entrench the incumbent's market position. [27][9][10]
• Anthropic withheld its most powerful unreleased model after testers called it a 'super weapon,' while simultaneously deploying Fable 5 and Mythos 5 commercially until the government forced them offline — suggesting the internal threshold for withholding a model differs from what Anthropic considered safe for public deployment. [20][15]