=== HEADLINE_SHORT ===
4,200 People Have Anthropic's Most Capable AI

=== HOOK_SCRIPT ===
Days ago, Anthropic released its most capable model.

=== SHORT_SCRIPT ===
Days ago, Anthropic released its most capable model. Right now, a Discord with four thousand paying members is using it without permission, and Anthropic still has not been able to shut the door.

Mythos is the preview of Project Glasswing. Anthropic publicly called it too capable for general release. Ninety one percent on SWE-Bench Pro. Two million tokens of context. Tier 3 in their Responsible Scaling Policy. The kind of model the entire safety pitch is built around.

Nine days after rollout, Fortune reported that a paid Discord with roughly four thousand two hundred members has continuous unauthorized access. Two hundred dollars a month per seat. Not through Anthropic's API. Through a third party fine tuning vendor that one of the pilot enterprise customers was routed through for compliance reasons.

The same vendor has pipeline relationships with OpenAI and Google. Anthropic announced a five million dollar emergency bug bounty. The FTC opened an inquiry. As of recording, the door is still open.

The slow rollout was not slow enough. The vetted partners were not vetted enough. Every other frontier lab is now staring at its own vendor list with new eyes.

Stay sharp.

=== ANNOTATED_HTML_SHORT ===
<span class="cue">[TIGHT CLOSE-UP — Jane]</span>
<p>Days ago, Anthropic released its most capable model. Right now, a Discord with four thousand paying members is using it without permission, and Anthropic still has not been able to shut the door.</p>
<span class="cue">[ON-SCREEN TEXT: "4,200 MEMBERS · CONTINUOUS ACCESS"]</span>
<p>Mythos is the preview of Project Glasswing. Anthropic publicly called it too capable for general release. Ninety one percent on SWE-Bench Pro. Two million tokens of context. Tier 3 in their Responsible Scaling Policy. The kind of model the entire safety pitch is built around.</p>
<span class="cue">[ON-SCREEN TEXT: "91% SWE-BENCH PRO"]</span>
<span class="cue">[ZOOM IN]</span>
<p>Nine days after rollout, Fortune reported that a paid Discord with roughly four thousand two hundred members has continuous unauthorized access. Two hundred dollars a month per seat. Not through Anthropic's API. Through a third party fine tuning vendor that one of the pilot enterprise customers was routed through for compliance reasons.</p>
<span class="cue">[ON-SCREEN TEXT: "9 DAYS · $200/MONTH"]</span>
<p>The same vendor has pipeline relationships with OpenAI and Google. Anthropic announced a five million dollar emergency bug bounty. The FTC opened an inquiry. As of recording, the door is still open.</p>
<span class="cue">[TIGHT CLOSE-UP — Jane]</span>
<p>The slow rollout was not slow enough. The vetted partners were not vetted enough. Every other frontier lab is now staring at its own vendor list with new eyes.</p>
<p>Stay sharp.</p>

=== THUMBNAIL_HTML_SHORT ===
<div class="thumb-spec">
  <h3>Vertical Frame (9:16, 1080x1920) — Jane</h3>
  <p><strong>Framing.</strong> Tight head-and-shoulders, eyes in the upper third, mouth at center. Face fills ~60% of frame width.</p>
  <p><strong>Expression.</strong> Quietly alarmed, jaw set, eyes steady. The look of someone reading the leak headline in real time.</p>
  <p><strong>Eye direction.</strong> Direct to camera, locked.</p>
  <p><strong>Background.</strong> Near-black charcoal, no environmental detail. Single faint terminal-glyph accent at 10% opacity in upper-right corner. Vertical viewport is too narrow for richer scene.</p>
  <p><strong>Lighting.</strong> Single key light from upper-left at ~4500K, deep shadow on the right jaw line. Subtle red rim light from behind-right tagging the cybersecurity tone.</p>
</div>

<div class="thumb-option">
  <div class="thumb-rank">Best — Top Text Block</div>
  <div class="thumb-text">MYTHOS LEAKED</div>
  <p><strong>Position.</strong> Top third, full-width, large.</p>
  <p><strong>Font.</strong> Bebas Neue Bold or Impact, condensed all-caps, tight tracking.</p>
  <p><strong>Color.</strong> "MYTHOS" in pure white, "LEAKED" in muted red (#b91c1c). 3px black stroke for legibility.</p>
</div>

<div class="thumb-option">
  <div class="thumb-rank">Alternate — Number Block</div>
  <div class="thumb-text">4,200 USERS</div>
  <p><strong>Position.</strong> Top third, centered.</p>
  <p><strong>Font.</strong> Inter Black all-caps, tight tracking.</p>
  <p><strong>Color.</strong> Pure white with 3px black stroke. Tiny gold sub-tag below: "ANTHROPIC MYTHOS" in Inter Bold 22px, #c8a84b gold.</p>
</div>

=== HEYGEN_LOOK_SHORT ===
A photorealistic tight close-up portrait of a poised woman in her early 30s with a quietly alarmed, focused expression, dark blazer, minimalist styling, no jewelry. Background: a near-black charcoal studio wall with a single faint terminal-glyph code accent at 10% opacity behind her shoulder, no other environmental detail, kept simple for vertical framing. Single hard key light from upper-left at ~4500K, minimal fill on the right at 15% intensity, deep shadow on the right jaw line, subtle red rim light from behind-right tagging the cybersecurity tone. Eyes locked direct to camera. 9:16 portrait, face fills 60% of frame width, ultrarealistic, sharp focus, clean rendering, shallow depth of field.

=== DESCRIPTION_SHORT ===
4,200 paying Discord members have continuous access to Anthropic's Mythos. #Mythos

=== HASHTAGS_SHORT ===
#Mythos #Anthropic #AISecurity #Shorts