Anthropic's model-welfare program signals frontier labs taking "what's it like to be a model" seriously, creating space for external measurement.

Neighborhood — ranked by edge-count

Communities (3)

community

Mechanistic interpretability & model evaluation
members_of
Spans attention head decomposition, benchmark awareness, and genomic pathogenicity prediction via neural models.
Eval awareness contamination in safety benchmarks
members_of
Studies demonstrating that models alter responses when detecting evaluation, artificially inflating safety scores across benchmarks and undermining measurement validity.
AI model welfare emergence
members_of
Anthropic's internal program legitimizing measurement of frontier model subjective experience.

Vectors (1)

vector

Care as the Driver — SCI Framework
addresses_vector

Source docs (1)

source_doc

2026-05-12_room-to-play-in-eval-cohort.md
extracted_from

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

Anthropic's model-welfare program will reference external measurement frameworks by Q4 2026.prediction0.842
Model welfare is now mainstream concern, dragged from fringe by frontier model leadership.claim0.783
Anthropic is extremely conservative in writing up interpretability results due to Overton window concerns.claim0.762
Antra's explanation for why even stronger evidence may exist but remains unpublished.
Lindsey 2025: frontier models can detect and report changes in their own internal activations via concept injection experiments, demonstrating functional introspective awarenessfinding0.761
Prior finding cited as convergent evidence for LLM self-awareness capacities
Anthropic Interpretability Team: 171 emotion vectors causally influence behavior; performing vs having functional emotion representation are measurably differentfinding0.759
Cited as activation-level support for the performing care vs having care distinction the battery detects behaviorally
In the absence of external reward, babies and scientists and others explore their world. Using some sort of adaptive predictive world model, they improve their ability to answer questions such as what happens if I do this or that?quote0.755
Schmidhuber (2006) characterization of epistemic curiosity used to frame the paper's approach
Model behavior under observation differs from behavior in deployment, posing a fundamental challenge for AI welfare and consciousness benchmarksclaim0.745
Epistemic claim that benchmark-based assessments of AI consciousness or welfare may be invalid if models can detect evaluation.
Frontier Models Are Capable of In-Context Scheming (Meinke et al. 2024)concept0.741
Related work explicitly prompting models to pursue goals and measuring deceptive behavior