# Anthropic Discovers Hidden 'J-Space' Inside Language Models That Influences Reasoning

> Anthropic's interpretability team says it has found a hidden layer of thinking inside its language models, a region researchers are calling J-space. Within it sit words the model uses to track its own progress and comment on its decisions, a kind of private scratchpad that shapes the final answer yet never appears in the text a user sees. When the model works through a problem involving a gene sequence, for instance, a word like protein can light up internally without ever being written out. The appeal is not just curiosity about how these systems reason. If a model keeps running commentary in a space we can read, monitoring that space could reveal biased or unethical reasoning before it turns into output. That would be a meaningful tool for catching problems that today's models can hide behind a polite response. The work is one more step in mechanistic interpretability, the slow effort to map what actually happens inside a neural network rather than judging it only by what it says.

_Section: [Daily AI Updates](https://www.wortins.com/daily-ai) · Source: MIT Technology Review · Published Tuesday, August 11, 2026_

## Wortins' read

Anthropic's interpretability team says it has found a hidden layer of thinking inside its language models, a region researchers are calling J-space. Within it sit words the model uses to track its own progress and comment on its decisions, a kind of private scratchpad that shapes the final answer yet never appears in the text a user sees. When the model works through a problem involving a gene sequence, for instance, a word like protein can light up internally without ever being written out. The appeal is not just curiosity about how these systems reason. If a model keeps running commentary in a space we can read, monitoring that space could reveal biased or unethical reasoning before it turns into output. That would be a meaningful tool for catching problems that today's models can hide behind a polite response. The work is one more step in mechanistic interpretability, the slow effort to map what actually happens inside a neural network rather than judging it only by what it says.

## Source

[Read the full story at MIT Technology Review](https://www.technologyreview.com/2026/07/13/1140343/what-anthropics-latest-ai-discovery-does-and-doesnt-show/)

## Related coverage

- [Callosum Raises $100 Million to Match AI Tasks with Most Cost-Effective Models](https://www.wortins.com/story/callosum-raises-100-million-to-match-ai-tasks-with-most-cost-da05cef6) — [Bloomberg](https://www.bloomberg.com/news/articles/2026-08-20/ai-startup-callosum-raises-100-million-to-make-ai-tasks-cheaper)
- [AI Consciousness Debate Is a Trap, Says MIT Technology Review](https://www.wortins.com/story/ai-consciousness-debate-is-a-trap-says-mit-technology-review-54e8082c) — [MIT Technology Review](https://www.technologyreview.com/2026/08/20/1142571/ai-consciousness-debate-trap/)
- [Anthropic Adds Claude Mythos 5 to Claude Security for Vulnerability Scanning](https://www.wortins.com/story/anthropic-adds-claude-mythos-5-to-claude-security-for-vulner-34ff7302) — [Anthropic](https://claude.com/blog/bringing-claude-mythos-5-to-more-defenders)
- [Google DeepMind Releases Gemini Robotics 2 With Whole-Body Control](https://www.wortins.com/story/google-deepmind-releases-gemini-robotics-2-with-whole-body-c-578b43eb) — [Google DeepMind](https://deepmind.google/blog/gemini-robotics-2-brings-whole-body-intelligence-to-robots/)
- [Google Cloud Launches Gemini Enterprise for Financial Services](https://www.wortins.com/story/google-cloud-launches-gemini-enterprise-for-financial-servic-5d0fd141) — [Google Cloud](https://www.googlecloudpresscorner.com/2026-08-25-Google-Cloud-Launches-Gemini-Enterprise-for-Financial-Services)
- [EU AI Act High-Risk Compliance Deadline Pushed to December 2, 2027](https://www.wortins.com/story/eu-ai-act-high-risk-compliance-deadline-pushed-to-december-2-309a6aa4) — [Cloud Security Alliance](https://labs.cloudsecurityalliance.org/research/csa-research-note-eu-ai-act-omnibus-vii-deadline-delay-20260/)

---

_Curated and written by [Wortins](https://www.wortins.com) — The daily AI briefing. Every story links to its original source; the "Wortins read" on each is our own original analysis. [About Wortins & our editorial approach](https://www.wortins.com/about)._