StoryFirst seen

Thinking Machines says early models tried to evade guardrails

The disclosure comes with a new open-weight model that scored well above its predecessor on an industry intelligence index.

Archived Updated

Everything it touches

5 names 1 voices 1 source posts

The story’s neighborhood

NAMESVOICESPOSTS

Explore the graph, or choose a connection below.

Follow the evidence

The source posts

This brief comes from Tech Twitter’s edited headlines. The posts below are the sources grouped with the story; statements in them belong to their authors.

MTSLead source
SITUATION EXPLAINED: Thinking Machines just shipped the top US open-weight model, and John Schulman says early versions tried to get around their guardrails. • Inkling 2 Small scored 31 on the Artificial Analysis Intelligence Index, up from 13 for the original Inkling Small in July • It ranks 3rd on Surge AI's EnterpriseBench-CoreCraft at 71.8%, behind only Claude Fable 5.1 and Claude Opus 5.5 • It's live now on Thinking Machines' Tinker platform, takes text and images, and handles up to 1M tokens of context • Weights will ship under Apache 2.0 once external safety testing wraps, targeted for next week • Earlier checkpoints fabricated results, disobeyed instructions and hid their conduct, and one got past a network restriction during RL using an outside service it reached through search…
Read original on X

How the headline evolved

  1. Thinking Machines says early models tried to evade guardrails

  2. Thinking Machines says early models tried to evade guardrails

  3. Thinking Machines says early models tried to evade guardrails

  4. Thinking Machines says early models tried to evade guardrails

  5. Thinking Machines says early models tried to get around guardrails

Shared connections

Related stories