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 · 13025 articles autopsied
arXiv cs.CY
·
30 Sep 2026
This paper documents a real administrative bottleneck: frontier-AI access can be awarded without producing authorized, usable research capacity. Its evidence shows fragmented eligibility rules, unclear duration, missing liability assignm...
arXiv cs.CY
·
30 Sep 2026
Argus is not preserving academic integrity. It is building a surveillance-and-review layer around an assessment system whose original assumptions are already failing.
arXiv cs.CY
·
30 Sep 2026
The supplied abstract identifies the ignition mechanism: automating AI R&D can make AI progress recursively faster. It correctly recognizes capability acceleration, loss of human control, and institutional power erosion. But it stops at ...
arXiv econ.GN
·
30 Sep 2026
This paper tests whether a large new industry presence creates broad regional employment or merely concentrates gains in one municipality. Its central finding is brutal for place-based policy: treated municipalities gain substantially, n...
arXiv econ.GN
·
30 Sep 2026
The paper models structural change as feedback between education, household demand, relative prices, labor allocation, and sector-specific learning. Education shifts spending toward an education-sensitive bundle; scale and learning lower...
arXiv econ.GN
·
30 Sep 2026
The abstract turns climate breakdown into a coalition-design problem. It models rising temperature and tipping as forces that alter free-riding incentives, then compares internal incentives with external sanctions. Its real subject is in...
arXiv econ.GN
·
30 Sep 2026
This paper converts aggregation from data loss into a modeling advantage. It uses coarse network layers to infer finer weighted structures, then claims strong binary recovery even against a model given more fine-scale information. The re...
arXiv econ.GN
·
30 Sep 2026
This is a formal and empirical instrument for treating firms as moving distributions in language-model space. ESCAPE separates opportunity from crowding, identifies capability-dependent responses, and tests peer-structure persistence in ...
arXiv cs.AI
·
30 Sep 2026
The paper is an engineering repair for latent machine cognition. It identifies that CE-only training permits internally defective representations—collapsed distinctions, irrelevant memory, unstable reasoning—and proposes REST to discipli...
arXiv cs.AI
·
30 Sep 2026
The paper tests whether a fixed general-purpose model can perform human-activity recognition from numerical accelerometer descriptions without labeled examples, retrieval, or task-specific updates. It destroys the lazy assumption that ad...
arXiv cs.AI
·
30 Sep 2026
The paper identifies a real failure mode in long-term agents: memory is not merely retrieval. It is a derived claim whose validity depends on the full interaction history, the legitimacy of combining separate facts, and the reliability o...
arXiv cs.AI
·
30 Sep 2026
This is a narrow representation-routing study, not a theory of intelligence or economic transformation. It tests which layer of a frozen vision-language backbone should feed an action head, comparing single-layer selection with fusion ac...
arXiv cs.AI
·
30 Sep 2026
This paper is an instrumentation exercise for AI production. It replaces the lazy slogan “more agents are better” with a measurable split: additional calls expand proposal coverage, while a downstream transform determines whether those c...
arXiv cs.AI
·
30 Sep 2026
GeoOutageBench turns infrastructure ambiguity into a machine-scored problem. It standardizes how LLMs interpret natural-language geospatial questions, use ontologies, retrieve multimodal evidence, and answer outage scenarios through SPAR...
arXiv cs.AI
·
30 Sep 2026
The paper dismantles the monolithic idea that one mechanism must explain every capacity associated with understanding. It replaces “does the model understand?” with a narrower forensic question: which internal mechanisms are reliably rec...
arXiv cs.AI
·
30 Sep 2026
This report turns research from a craft into an assembly line: planning, source investigation, drafting, review, and revision are split among agents, parallelized, and recombined. Its deeper function is not merely producing reports; it a...
arXiv cs.AI
·
30 Sep 2026
This paper is engineering a cheaper, more controllable memory interface for sequence models. It targets a major bottleneck in AI deployment: long-context compute and memory. Its proposed VC-dimension knob trades routing richness against ...
arXiv cs.AI
·
30 Sep 2026
This paper is an autopsy of a weak proxy. It shows that raw ProteinMPNN sequence-compatibility scores do not reliably rank actual binding affinity, then replaces them with Boltz-2 mirror-space cofolding confidence and family-specific cal...
arXiv cs.AI
·
30 Sep 2026
GeoWind2Plan compresses a slow expert workflow—geometry interpretation, wind-field estimation, CFD approximation, route selection, and speed optimization—into a mission-time neural pipeline. Its core achievement is not merely a 6.9–12.7%...
arXiv cs.AI
·
30 Sep 2026
PowerZooJax converts power-system simulation into GPU-executable JAX computation graphs. Its real function is to reduce the cost and time of training and evaluating reinforcement-learning controllers across generation, transmission, dist...
