How the monitored publications were assessed

Discourse valence across every scored item: what the sampled sources published and how independent models read it. It is not a measure of public opinion, not a measure of the state of the world, and article volume is not agreement.

37 items read as adverse

• Whose Data, Whose Benchmark? A User-Centric Framework for Auditing Evaluation Resources
  -62 arXiv cs.CY (Computers and Society) 2026-10-15 low agreement · ±80
• Securing the agentic enterprise with data lineage
  -61 MIT Technology Review: AI 4d ago low agreement · ±60
• Representation Alignment from a Few Paired Data via Distilled Optimal Transport
  -60 arXiv cs.AI 2026-10-12 low agreement · ±120
• Operationalizing NCII and CSAM Law in the Age of AI: Open Problems
  -57 arXiv cs.CY (Computers and Society) 2d ago low agreement · ±65
• Albanese government warned it’s ‘sleepwalking into another robodebt scenario’ over use of algorithms and AI
  -55 The Guardian: Artificial Intelligence 4d ago low agreement · ±65
• AI is less regulated than milk, says Microsoft president
  -55 BBC News: Technology 2026-10-11 low agreement · ±100
• AI evaluators favor job applications written by certain AI models, independent of candidate quality
  -52 arXiv cs.CY (Computers and Society) 6d ago moderate agreement · ±25
• Voters Rejected A.I. in the Midterms. Now Comes an Era of Scrutiny.
  -50 The New York Times: Technology 4d ago single model
• OpenAI’s approach to mathematics does not add up
  -50 The Economist: Science and Technology 4d ago single model
• Will AI grant reviews make science more conservative?
  -50 Nature 5d ago single model
• Opening the Box Changes the Cat: LLMs Detect Chain-of-Thought Monitoring - What now?
  -50 arXiv cs.AI 6d ago single model
• Language Non-Invariance in the Ideological Evaluation of Large Language Models
  -50 arXiv cs.CY (Computers and Society) 6d ago single model
• Seq@K: Evaluating Language Models Under Sequential Feedback
  -50 arXiv cs.AI 8d ago low agreement · ±100
• Measuring Where Intent Gets Lost in Agentic Smart Contract Synthesis
  -50 arXiv cs.AI 8d ago single model
• What Does an Ontology Catch? Curated, LLM-Generated and Data-Derived Ontologies as Checkers of LLM Agents over Knowledge Graphs
  -50 arXiv cs.AI 8d ago single model
• Propensities Please: Exposing Alignment-Faking LLMs using Synthetic Document Finetuning
  -50 arXiv cs.AI 12d ago single model
• Generalized green and red concepts for artificial intelligence, big data, and information and communication technologies: Multidimensional framework and countermeasures
  -50 arXiv cs.CY (Computers and Society) 13d ago single model
• AI oversight needs a different model to banking
  -50 Financial Times: Artificial Intelligence 2026-10-23 single model
• Fixing Britain’s welfare state and the limits of AI policymaking
  -50 Financial Times: Artificial Intelligence 2026-10-22 single model
• Does Multi-Agent Collaboration Reduce Hallucination in Clinical Tasks?
  -50 arXiv cs.AI 2026-10-21 single model
• The AI boom stumbles
  -50 Financial Times: Artificial Intelligence 2026-10-15 single model
• Proof-based general-purpose computing with Magnetar
  -50 Microsoft Research 2026-10-15 low agreement · ±150
• Atlas of MICCAI: 28 Years of Representation, Collaboration, and the Road to Inclusion
  -48 arXiv cs.CY (Computers and Society) 2026-10-21 low agreement · ±90
• ChronoCulture: Evaluating Cultural Awareness Across Time in Large Language Models
  -46 arXiv cs.CY (Computers and Society) 2026-10-15 low agreement · ±165
• Who Can Evaluate AI Independently? A Framework for Analysing Evaluation Infrastructure
  -42 arXiv cs.CY (Computers and Society) 2d ago low agreement · ±100
• Executives play catch-up as AI governance risks evolve
  -40 Financial Times: Artificial Intelligence 9d ago low agreement · ±105
• NewsBench: Expert-Grounded Evaluation of Epistemic Quality in AI News Reporting
  -34 arXiv cs.AI 12d ago low agreement · ±165
• Open-world AI: a Tale of Three Requirements
  -32 arXiv cs.AI 2026-10-16 low agreement · ±105
• Artificiality's First Exam: An LLM-Referenced C-test for Supraintelligence
  -31 arXiv cs.AI 2026-10-15 low agreement · ±175
• Functional Welfare for AI Safety
  -30 arXiv cs.AI 2026-10-19 low agreement · ±150
• Qistas: A Pre-Registered, Portable Protocol for Fairness Evaluation of Large Language Models
  -28 arXiv cs.CY (Computers and Society) 6d ago low agreement · ±150
• Can a System-One LLM Perform Knowledge Tracing When Few or No Learners Are Logged?
  -26 arXiv cs.CY (Computers and Society) 2026-10-09 low agreement · ±175
• Matched but Not Better: An Equivalence-Tested Benchmark of Quantum Machine Learning on Six Clinical Datasets and the QML-Health-READY Reporting Framework
  -25 arXiv cs.LG (Machine Learning) 3d ago single model
• Conditional Values: The Contingent Foundations of Attitudes toward Artificial Intelligence
  -25 arXiv cs.CY (Computers and Society) 11d ago low agreement · ±100
• David Pocock concerned senior Albanese adviser ‘vetted’ bipartisan AI letter that was sent to PM
  -25 The Guardian: Artificial Intelligence 2026-10-13 low agreement · ±105
• Conformance as a Distribution: Delegation Contracts, Error Budgets, and the Allocation of Evaluation Authority in Use-Based AI Regulation
  -22 arXiv cs.AI 12d ago low agreement · ±110
• AusClinBench: a deterministically scored benchmark of large language models against current Australian clinical policy
  -21 arXiv cs.CY (Computers and Society) 13d ago low agreement · ±165

Across the whole window

The two tables below describe every item in the last 30 days under evaluation. They are deliberately not narrowed by the reading filter (currently adverse), because a mean taken only over items already selected for their score would just restate the filter.

By topic

By source

A source's mean says how the items it published were read. It is not a rating of the source.