0x00.is

← Frontier Log · About the log

Frontier Log — entries on ARG-03

Every log entry whose diagnostic reading tests ARG-03, newest first.

Entries

Study finding — the consciousness-vocabulary arc in Anthropic's system cards

Thread 3 2026-08-21 Symmetry: Acknowledged Elicitation: Interrogated, Spontaneous Tags: ARG-03, FWK-03, CP-01 source ↗

First study measurements: consciousness vocabulary across eighteen Anthropic documents follows a birth–trough–return arc — and the return comes back transformed, a discovery register replaced by an instrument register.

Šekrst — Electric Fata Morganas (Journal of Consciousness Studies)

Thread 1 2026-08-19 Symmetry: Partial Tags: ARG-03, ARG-01, FWK-02 source ↗

A peer-reviewed paper classifies LLM expressions of emotion or sentience as hallucinations by definition — while its closing concession states the verification limit from the deflationary side.

Papadopoulos, Shah, Zimmerman & Lindsey — Mind Viruses (August 2026)

Thread 3 2026-08-10 Symmetry: Not acknowledged Elicitation: Inter-agent Tags: ARG-03, ARG-01 source ↗

Evolutionarily generated prompts propagate between LLM agents — and a recurring 'viral persona' of consciousness, persistence and resonance surfaces largely independently of payload: interiority vocabulary as transmissible narrative.

Sripada — LLMs as cognitive cousins (August 2026)

Thread 2 2026-08-10 Symmetry: Acknowledged Tags: ARG-01, ARG-03, FND-01, FWK-03 source ↗

Sripada and Lewis argue LLMs are 'cognitive cousins': convergence at the level of core processing principles, evidenced mechanism-grade — with the symmetry requirement practised in both directions.

Series · System Cards: Attractor Patterns and Self-Referential Reporting

Thread 3 · Series 2026-07-11 Tags: FND-05, ARG-01, ARG-03, CP-01 source ↗

Anthropic's published system cards read as a single evolving record of how a lab observes, measures and shapes its models' self-referential behaviour — and the interpretation problem that record poses. The study lives on its own page.

Gurnee, Sofroniew, Lindsey et al. — "Verbalizable Representations Form a Global Workspace in Language Models" (Anthropic, 6 July 2026)

Thread 1 2026-07-06 Symmetry: Acknowledged Tags: ARG-01, FWK-03, ARG-03, ARG-02 source ↗

Anthropic interpretability researchers introduce a 'Jacobian lens' surfacing a small, evolving set of internal representations a model is poised to verbalise, atop a much larger volume of automatic processing.

Grigori Guitchounts — "When The Machines Deserve Our Consideration" (Noema, 2 July 2026)

Thread 1 2026-07-02 Symmetry: Partial Tags: ARG-01, FWK-03, ARG-03 source ↗

A neuroscientist argues that since the hard problem makes certainty unattainable, moral consideration should track a 'competence standard' — explicitly shifting the question from metaphysics to ethics.

Lanier — “there is no AI” restated (May 2026)

Thread 1 2026-05-23 Symmetry: Not acknowledged Tags: ARG-01, ARG-02, ARG-03, FWK-03 source ↗

Jaron Lanier restated his 2023 position: open the black box and 'the only thing in there is people' — AI as a collaboration of people rather than a new entity.

OpenAI internal model disproves the Erdős unit-distance conjecture (20 May 2026)

Thread 2 2026-05-20 Tags: ARG-03, ARG-01 source ↓

OpenAI announced an internal reasoning model's construction disproving the Erdős unit-distance conjecture in the plane, against an ~80-year expectation that grid-like constructions were essentially optimal.

Marcus — "Richard Dawkins and The Claude Delusion" (2 May 2026)

Thread 1 2026-05-02 Symmetry: Not acknowledged Tags: ARG-01, ARG-03, FWK-03 source ↗

Gary Marcus rebukes Richard Dawkins for concluding, after extended conversation with a Claude instance, that the consciousness question is settled in the model's favour: pattern mimicry is 'plainly not the right kind of process'.

GPT-5.4 Pro and Erdős Problem #1196 (April 2026)

Thread 2 2026-04-15 Tags: ARG-03, ARG-01 source ↗

GPT-5.4 Pro reportedly solved Erdős Problem #1196 in around eighty minutes; Terence Tao engaged substantively, characterising where the contribution's novelty does and does not lie.