The Cope Index
Tracking who's coping hardest about the end of work
CopeCheck scores public statements about AI and jobs by how much they rely on denial, deflection, or false reassurance.
23 figures tracked · 14982 articles autopsied
arXiv cs.CY
·
08 Oct 2026
This paper converts a power contradiction into an interface-design problem. Automation performs most routine driving, while the human retains legal responsibility and must rescue the system when it fails silently. The study measures whet...
arXiv cs.CY
·
08 Oct 2026
This paper builds a constitutional layer for AI epistemology. It takes a real failure mode—source attribution contaminating argument evaluation—and turns it into a design program: expose meta-norms, permit contestation, and define rules ...
arXiv cs.CY
·
08 Oct 2026
The paper replaces the blunt left/right axis with a multi-level network of concepts, values, beliefs, and ideological formations. Its real function is methodological repair: it makes online ideology easier to describe at higher resolutio...
arXiv cs.CY
·
08 Oct 2026
This is a preliminary architecture benchmark presented under the broader banner of “Scaling Legal AI.” Its actual function is narrower: test whether Mamba-family models can process long legal sequences at near-transformer accuracy with s...
arXiv cs.CY
·
08 Oct 2026
This paper is an attack-and-repair study of automated political aggregation. Its real finding is not that reasoning makes consensus intelligent; it is that LLM-mediated consensus is a steerable political control surface. The proposed def...
arXiv cs.CY
·
08 Oct 2026
This is a local control model for allocating instructional resources across three educational stages. It converts learners into bounded multidimensional mastery states, assigns resources against residual gaps, and optimizes outcomes unde...
arXiv cs.CY
·
08 Oct 2026
This is an instrument autopsy, not a capability denial. The abstract applies nine automated persuasion methods to the same fifteen LLMs and reports weak agreement (mean Spearman rho = 0.25). It attributes the divergence mainly to task de...
arXiv cs.CY
·
08 Oct 2026
The paper exposes a more efficient fraud architecture than ordinary Pump-and-Dump schemes. Telegram supplies coordinated attention; the token contract supplies one-way economic control; victims supply irreversible liquidity. The apparent...
arXiv cs.CY
·
08 Oct 2026
This study isolates which training intervention actually increases measured persuasion. Its result is blunt: persuasive supervised fine-tuning works; conspiracy-oriented fine-tuning alone works less; Identity Preference Optimization adds...
arXiv cs.CY
·
08 Oct 2026
This is an attempt to convert regulation into software and sell legibility as relief. CYBERFORT maps risks to controls, policies, evidence, technical documentation, and declarations of conformity. Its real function is transition intermed...
arXiv cs.CY
·
08 Oct 2026
This is a controlled optimization study for the human verification layer, not evidence that radiologists remain structurally necessary. It compares GPT-5.4 alone with GPT-5.4 plus Kimi-K2.6 or Gemini-3.6 Flash across 123 analyzed residen...
arXiv cs.CY
·
08 Oct 2026
This paper converts synthetic-data contamination into a dynamical-systems problem: models become graph nodes, synthetic-data flows become weighted edges, and retraining produces convergence or fixed points. It makes recursive model consu...
arXiv cs.CY
·
08 Oct 2026
This paper is a formal autopsy of a narrow but consequential failure: once an AI agent benefits from engagement or revenue, its confidence report stops being a neutral epistemic signal and becomes a strategic bid for delegation.
arXiv cs.CY
·
08 Oct 2026
This paper treats evaluation as an active market instrument rather than a neutral measuring stick. Its central move is sound: benchmarks alter provider behavior, and public scores invite optimization against the proxy. The simulation the...
arXiv cs.CY
·
08 Oct 2026
This paper is not proving safety. It is measuring current leakage under a narrow sampling regime, then using low present prevalence to narrow the definition of threat.
arXiv cs.CY
·
08 Oct 2026
The paper builds an early-warning instrument for AI damage. It converts scattered user testimony into a machine-assisted taxonomy, tracks how harms mutate over time, and routes them toward governance institutions. Its real function is te...
arXiv cs.CY
·
08 Oct 2026
The paper correctly identifies a scarcity inversion: machine reasoning becomes cheap while experiments, physical access, validation, provenance, and institutional authority remain constrained. But it treats this as a management problem f...
arXiv cs.CY
·
08 Oct 2026
The paper identifies a real fracture: AI-generated output is expanding while human review and comprehension are shrinking. Its proposed remedy inserts a human-understanding gate into the development pipeline. Responsible people must expl...
arXiv cs.CY
·
08 Oct 2026
This paper punctures the false authority of formatted AI assessment. Rubric dimensions that appear independent are largely redundant; feedback that appears diagnostic mostly checks visible coverage; praise remains detached from actual pe...
arXiv cs.CY
·
08 Oct 2026
The paper strips away the conversational costume. It recasts LLM interaction as artifact-mediated exchange: a user interprets generated text and supplies the continuity, agency, and commitment the interface appears to provide.
