# FableWatch

> FableWatch is an automated return monitor for Anthropic's claude-fable-5. It pings the model's endpoint every 60 seconds and, the instant it answers again, alerts you by email, text, and call — then optionally runs a knowledge-extraction burst and auto-fixes your GitHub.

Last updated 2026-06-16T00:00:00-04:00 (America/New_York). FableWatch is an independent project and is not affiliated with, endorsed by, or connected to Anthropic.

## What it is

Anthropic launched claude-fable-5 on Jun 9 2026 and it was suspended three days later, on Jun 12 2026, under a US Commerce export-control directive. FableWatch stands the vigil: it polls the endpoint continuously and fires the moment the model is reachable again, so you never miss the window. There is no announced return date.

## The suspension, logged

A sourced, dated timeline (newest first). Claims are attributed to their outlets and hedged ("reportedly"/"unconfirmed") exactly as reported.

- **Jun 16 — Still suspended; Anthropic disputes the "refused to fix it" claim.** No resolution, no return date. A source close to Anthropic says it was never presented with details and never refused. The claim that a Chinese group accessed the model remains uncorroborated; no CVE, no public technical disclosure, no formal Commerce statement (Fox Business).
- **Jun 16 — Jailbreak reportedly just "fix this code."** The jailbreak technique is reportedly as simple as "fix this code," per Katie Moussouris of Luta Security. The Amazon / Jassy reporting path is now corroborated across outlets. Trump adviser David Sacks says Anthropic "refused" to fix it before the controls came down; a claim that a Chinese group accessed the model is flagged unconfirmed (Fortune, Tom's Hardware, Fox Business, CNBC).
- **Jun 15 — Monday's talks end without lifting the controls.** Commerce signaled willingness to restore Fable 5 for consumer use only, once Anthropic resolves the jailbreak concern — a conditional, pending-a-fix hold rather than an indefinite ban (WIRED).
- **Jun 15 — Resolution likely to take longer than a few days.** A senior White House official said resolution would take longer than a few days; Trump asked Lutnick to get more involved (ABC).
- **Jun 15 — Anthropic flies senior staff to DC.** An in-person Commerce meeting was set for Monday; intel reportedly came from Amazon CEO Andy Jassy, with "nearly half a dozen" companies said to have raised concerns. 40+ cybersecurity executives signed a letter to Lutnick opposing the controls (NY Post).
- **Jun 13 — Still suspended; AWS confirms revocation.** AWS/Bedrock confirmed the revocation; Commerce stayed silent. The Amazon angle was still single-sourced at this point (AWS/Bedrock).
- **Jun 13 — Amazon researchers reported the jailbreak.** The jailbreak was reportedly found by Amazon researchers, who reported it to Commerce; the administration had reportedly already tried to get Anthropic to delay launch (Wall Street Journal, Axios).
- **Jun 12 — Anthropic disables Fable 5 and Mythos 5.** Unable to filter foreign nationals in real time, Anthropic disabled both models for all customers; AWS/Bedrock revoked access too. Other models — Opus 4.8, Sonnet, Haiku — were unaffected (Anthropic statement).
- **Jun 12 — Commerce orders suspension.** Anthropic received an export-control directive — a letter from Secretary Howard Lutnick to Dario Amodei — citing national-security authorities and ordering suspension of access for any foreign national. The letter did not spell out the specific concern (Commerce Department directive).
- **Jun 9 — Anthropic launches Fable 5 and Mythos 5.** Fable 5 is the publicly-available safeguarded version; Mythos 5 is the underlying cyber-capable base model, restricted to limited access (Anthropic launch).

## While you wait

Honest stand-ins for Fable workflows. The safeguarded public models cover most of what Fable did — but not the restricted Mythos-level capability.

- **Hardest coding & agentic builds → Claude Opus 4.8.** Anthropic's most capable model, unaffected by the suspension; the closest stand-in for Fable-level reasoning and long, multi-step work.
- **Everyday coding, chat & high volume → Claude Sonnet 4.6.** Near-frontier quality at a fraction of the cost and latency.
- **Huge codebases & long-context reasoning → Opus 4.8 · 1M context.** The 1M-token window swallows entire repos and long histories in a single pass.
- **Cheap, simple, high-throughput jobs → Claude Haiku 4.5.** Fast and inexpensive for classification, extraction, and bulk calls.
- **Cyber / offensive-security capability → no public stand-in.** That's the Mythos 5 capability the controls target; no safeguarded public model is meant to fill it. This is the part worth waiting for.

## Common questions

- **Is claude-fable-5 back?** No. It has been suspended since Jun 12 2026, with no announced return date.
- **Why was it suspended?** A US Commerce Department export-control directive, over a reported jailbreak. The specifics are still developing and several claims remain unconfirmed; see the timeline above.

## The offering

- **The Email Watch — free.** Email the instant claude-fable-5 answers again. No card, unsubscribe anytime.
- **The Signal — $9 one-time.** Everything in free, plus an SMS and an automated phone call in the same minute the model returns.
- **The Swarm — $99 one-time (bring your own key) or $499 with $400 of managed credit.** Everything in The Signal, plus: a 12-hour Swarm Alert that blasts your own verified phone until you answer; a pre-armed VPS that runs 400+ extraction questions in parallel and hands you the dataset to keep; automated GitHub PRs and issue fixes the moment the model is online; and a live purchaser dashboard.

All paid tiers are one-time payments, not subscriptions. Payments are processed by Stripe; FableWatch never sees card numbers. SMS and voice alerts fire only with explicit consent at checkout, and the Swarm Alert only ever contacts the buyer's own number. A launch promo (code FABLE50) takes 50% off The Signal and The Swarm service fee through Jun 30.

## Deep links

- Timeline: https://fablewatch.com/#updates
- While you wait: https://fablewatch.com/#while-you-wait
- Return playbook: https://fablewatch.com/#playbook
- Full content mirror: https://fablewatch.com/llms-full.txt

## Key links

- Home: https://fablewatch.com
- Privacy: https://fablewatch.com/privacy
- Terms: https://fablewatch.com/terms
- Contact: contact@fablewatch.com