• Testing the global workspace at open frontier scale: the Jacobian lens on DeepSeek-V4-Flash and GLM-5.2

  A replication and extension of the Jacobian lens on two frontier-scale open models, separating workspace properties that transfer from those that vary by family, with a vendor-described same-base test of whether a second post-training run is visible in J-space.

  54 min read   ·   August 07, 2026

  2026   ·   language-model   mechanistic-interpretability   jacobian-lens   global-workspace   deepseek   glm   ·   LLM   MI

• AI and Productive Systems 3: Institutional Design for AI Agents

  Alignment, safety, and interpretability are necessary. AI governance also needs institutional design for ownership, work, compute, and accountability.

  10 min read   ·   May 19, 2026

  2026   ·   ai-productive-systems   alignment   constitutional-ai   interpretability   ai-safety   governance   ·   Society   Governance

• AI and Productive Systems 2: From Tool to Production Subject

  What changes when AI systems act as persistent, tool-using participants inside firms, markets, and institutions?

  11 min read   ·   May 18, 2026

  2026   ·   ai-productive-systems   production-subjects   workplace-governance   legal-identity   information-integrity   ·   Society   Governance

• AI and Productive Systems 1: Intelligence as Infrastructure

  AI looks like software at the interface. At scale, it depends on compute, energy, chips, labor, and institutions.

  11 min read   ·   May 17, 2026

  2026   ·   ai-productive-systems   compute   labor   governance   infrastructure   ·   Society   Governance

• Understanding Language Models 2: Stable Features and Identifiable Causal Structure

  A stable feature basis is the prerequisite for interpretable causal structure: first demand run-to-run consistency, then recover temporal and instantaneous relations.

  11 min read   ·   April 02, 2026

  2026   ·   language-model   mechanistic-interpretability   causal-representation-learning   sparse-autoencoder   identifiability   ·   LLM   MI   CRL