Evaluation and media provenance

Benchmarks, task datasets, attribution, metadata, interoperability and quality review

• articleHow do papers actually check that an interactive world model did what the action or camera input asked, and what are the four basic ways of doing it?
  The inspected primary papers use four different checks. (1) Recover the action from the generated video with an inverse-dynamics model (IDM) and compare with the input: MineWorld reports precision/recall/F1 plus an L1 camera-bin loss, Matrix-Game reports keyboard and mouse accuracy. (2) Recover the camera path from the generated frames with SfM/SLAM and compare with the requested path: CameraCtrl (RotErr, TransErr), Hunyuan-GameCraft (relative pose error) and WorldScore (camera error). (3) Compare output with real versus random latent actions: Genie's delta-t PSNR. (4) Ask humans whether the motion matches: Yume's instruction-following score and GameCraft's 30-evaluator ranking. Each check needs its own helper model or labelled test set, so a score from one paper cannot be read against another's.
• articleAre there public, reported tests that compare the same seed across different GPU vendors, drivers or library versions for image diffusion, and what differences do they report?
  No controlled cross-vendor or cross-driver pixel comparison for diffusion images was found in the sources searched. What exists are issue reports that touch the axes indirectly: CPU vs CUDA generator device on one GPU (diffusers 13298, Flux Schnell; maintainer: expected), MPS vs an RTX 5090 on the same graph where a second user could not reproduce the MPS failure on CUDA (ComfyUI 16879; the CUDA user also found CPU-fp16 vs GPU text encoder differ by MAD 0.6/255), an AMD gfx1100 INT8-attention failure producing noise that the default attention path avoids and that two users report fixed by comfy-kitchen 0.2.37 (ComfyUI 16711; library-version axis), and a video-model AMD thread where the reporter withdrew their first two causes (ComfyUI 15314). The maintainers' FAQ and A1111's option text claim CPU noise helps across vendors but show no test. The one cross-GPU-architecture paper found is about LLM token streams, not images.
• articleWhat public issue reports show the same seed and graph giving different images on one machine, what did the reporters measure, and what causes were identified or only suspected?
  Five public threads, all reported_execution by their authors, show run-to-run or configuration-induced differences on one machine: batch size 1 vs 2 (diffusers 2659, three GPUs, SD 1.5), SDPA attention run-to-run (diffusers 3061), batch-size-dependent scheduler noise (diffusers 6222, resolved by per-image generators), fp8 + LoRA + dynamic VRAM (ComfyUI 15365, ~5-7/255 mean absolute difference, bit-identical without the LoRA, mechanism a hypothesis) and a cold-vs-warm server ordering effect on Apple MPS (ComfyUI 16879, pixel-identical within each state). Causes named by maintainers or reporters are claims; only the measured differences are the reporters' observations, and none was independently reproduced by us.
• articleWhich textual signals make a reader (or a classifier) conclude that a narrator is unreliable, and how do the signals differ by what they depend on?
  The Handbook entries list concrete signals: contradiction with real-world fact, mismatch between an earlier report and the actual outcome, other characters' consistent disagreement, internal contradictions and verbal tics, conflicts between narrators, and paratextual types. They also stress that unreliability is a reader's hypothesis competing with other explanations (author error, genre, fictional-world rules, function), so a signal works only if it makes the narrator-based explanation the best one. An open 2025 LLM study operationalises three Hansen types and reports that detection works best for within-narration tics and poorly for signals needing comparison or outside knowledge.
• articleDoes the Physics-IQ Verified leaderboard support Odyssey's claim that Odyssey-3 Pro's 66.1 is the highest reported v2v score, and under what conditions was that number produced?
  Yes, narrowly. At repository commit d3f3de16 (observed 2026-10-22) and in the site records, 66.10 is the highest listed Verified video-to-video score, 0.63 above the next row (Cosmos3-Super best-of-8, 65.47), and it is not the highest for image-to-video (Odyssey-3 Pro best-of-8 54.69 vs Cosmos3-Super 55.10). The number is a single best-of-8 run with rewritten ("custom") prompts, self-submitted and merged by the repository maintainers without a stated reproduction; the highest v2v row without best-of-N is Odyssey-3 Pro prompt-enhanced at 63.37 ± 0.63, and the highest v2v row that lists any standard deviation is FLUX 3 [large] best-of-N at 64.35 ± 0.22. The repository rule for claiming SOTA asks for a 4-run standard deviation, which the 66.10 row lacks.
• articleRAP reports large planning gains from an LLM world model while Wang et al. find LLM state prediction unreliable; can the two be reconciled from the papers alone?
  The two papers do not contradict each other because they measure different things on different tasks with different models, but the papers alone cannot say why RAP works. RAP (LLaMA-33B, frozen) reports plan success of 0.88 at 4 steps and 0.42 at 6 steps with 20 MCTS iterations on Blocksworld. LLM-Sim (GPT-4) reports 49.7-77.1% best-case on changing transitions of a different set of games. Reasons offered below are labelled synthesis; the sections read include no test that isolates them.
• articleWhat do Adobe, Blackmagic and Avid publicly state about content credentials, usage terms and data handling for AI-generated or AI-processed material in the timeline?
  Adobe states in its Premiere pages that Generative Extend media is not used to train Adobe's model, that feedback uploads include the generated media (and optionally a 3-second low-resolution reference snippet), and that Premiere includes Content Credentials on export when the sequence contains Adobe AI-generated assets. Adobe's Generative AI User Guidelines (updated May 2026) forbid removing or altering Content Credentials and say prompts, inputs and results may be reviewed. No inspected Adobe Generative Extend page states that its output specifically receives Content Credentials. No Blackmagic or Avid document inspected mentions Content Credentials, C2PA, provenance or watermarks.
• articleDo documentary organisations, insurers or law firms say anything about E&O insurance and clearing or disclosing AI-generated material, and do the older fair-use and insurer documents mention AI at all?
  The Archival Producers Alliance guidelines (v.7, September 2024) and tool kit say to be transparent with legal counsel and E&O insurers, name risk "not covered by E&O insurance policy" as an item for attorneys, and tool kit question 8 asks whether GenAI materials will impact E&O; they also ask whether the distributor permits GenAI and what documentation it requires. Reed Smith (30 Sep 2026) says E&O coverage may turn on exclusions and gives an AI-assisted rights-clearance report that omits a licence as an exposure example. The Fair Use Statement page (Nov 2005) and both Beazley forms contain no AI terms in the text searched.
• articleWhich provenance marks (C2PA Content Credentials, SynthID, visible or other watermarks) do the image vendors say they apply, what do they say about removal, and where do their own pages disagree?
  OpenAI's help page (snapshot 2026-10-12) says images from ChatGPT, Codex or the API carry both Content Credentials and SynthID watermarks, with exceptions for models near deprecation. Google's Imagen docs say images carry a SynthID watermark by default (addWatermark true) and the Imagen 4 model page lists Content Credentials (C2PA) as not supported. Adobe applies C2PA metadata in different scopes depending on which Adobe page and date is read, and its guidelines forbid removing Content Credentials. Ideogram's terms prohibit removing watermarks, with no C2PA or SynthID text found; no provenance text was found in the Midjourney terms and help pages read.
• articleWhich image vendors offer a vendor-to-customer IP indemnity for generated output, on which surfaces, models and conditions, and which only state the customer's indemnity to the vendor?
  Three vendors publish a vendor-to-customer indemnity in the texts read, each narrow. Google Cloud's applies to unmodified Generated Output from listed Generative AI Indemnified Services (Imagen among generally available models) and not to free use, and covers training-data allegations separately. OpenAI's Service Terms extend its indemnity to API customers' use of Output with six exclusions, and exclude Beta Services. Adobe's Product Description lists Eligible Firefly Features, Surfaces and Export Events for customers whose agreement links that page, excluding non-Adobe models and beta features. The Midjourney and Ideogram terms read contain only the customer's indemnity of the vendor; OpenAI's consumer terms likewise.
• articleWhat does ComfyUI itself say a custom node is, where does it place the trust boundary, and which security reports does it refuse to treat as vulnerabilities?
  At ComfyUI commit 46aad33e (observed 2026-10-22) SECURITY.md states a local-first threat model: custom nodes are arbitrary Python code trusted like any installed software, anyone who can reach the URL is trusted, and issues needing --listen, a particular custom node, old PyTorch, or an ignored UI warning are out of scope. The loader code executes each custom node package's __init__.py in the server process at startup, which is consistent with that statement. The docs install page adds a warning to install only from trusted authors.
• articleHow does PhyGenBench / PhyGenEval check physical commonsense in text-to-video models, how well does it track humans, and what did it find about scaling and prompt engineering?
  PhyGenBench (Oct 2024) has 160 text prompts covering 27 physical laws in four domains (mechanics, optics, thermal, material properties) and PhyGenEval, a three-stage check that uses a VLM and an LLM-generated question set: key physical phenomena detection on retrieved key frames, physics order verification across first/key/last frames, and an overall naturalness grade. On 64 prompts and 8 models, three annotators rated 512 videos; PhyGenEval correlates 0.81 overall with humans and the authors show VideoPhy, VideoScore and DEVIL rate physically wrong videos as correct. The best model, Gen-3, reaches 0.51 and optics is easiest; the authors report scaling and prompt engineering do not fix dynamic phenomena.
• articleWhich general.* metadata keys does the GGUF spec define for identity, provenance and licence, which are required, and how far can an agent trust them to establish a model's licence or lineage?
  At ggml da8e2a54 the spec's 'Required' heading lists general.architecture, general.quantization_version and general.alignment, but the prose qualifies the latter two (quantization_version only if tensors are quantized; alignment defaults to 32). Identity and provenance keys (general.name, author, version, organization, basename, finetune, uuid, doi, url, repo_url, quantized_by, source.* and base_model.{id}.*) are recommended, not required. Licence is carried as general.license (an SPDX expression only, 'do not include any other information'), with optional general.license.name and general.license.link. These are self-declared strings written by whoever created the file; nothing in the spec says they are verified, so they are a lead for provenance, not licence evidence.
• articleCan an LLM serve as a judge of story or creative-writing quality, and how reliable and biased is it?
  Two 2023 papers disagree. Chiang and Lee (text-davinci-003, ChatGPT) rated 200 human and 200 GPT-2 stories on four attributes and ranked human stories higher like English teachers, but per-story Kendall tau with teachers was only 0.12-0.43 and T0/text-curie-001 did not separate them. Chakrabarty et al. had GPT-3.5, GPT-4 and Claude v1.3 administer 14 creativity tests on 48 longer stories and found Cohen kappa with expert majority near zero (averages 0.016, 0.035, -0.006). Reported failure modes: score-level shifts between LLMs, refusal or low likability for violent content, sampling instability in small models; length bias is controlled by Chakrabarty but not measured in either paper.
• articleWhat is Dramatron, how does it generate a script, and what did 15 theatre and film professionals actually use it for?
  Dramatron (DeepMind, Sept 2022) chains prompts hierarchically from a writer-supplied log line to title, characters, plot (scene beats), location descriptions and per-scene dialogue, using Chinchilla 70B with few-shot prompts; the writer can regenerate, continue or edit at every level. In a 2-hour paid study with 15 playwrights/screenwriters, writers used it mainly for inspiration, character/world alternatives and rough material, not finished scripts; five co-written scripts were staged at the 2022 Edmonton Fringe.
• articleDoes TikTok's own help-centre page say it reads Content Credentials on uploads and labels AI content automatically, and what do Meta's own pages say about its "AI info" label?
  In the 2026-06-27 snapshot TikTok may auto-apply an 'AI-generated' label when TikTok AI effects are used or uploaded content carries C2PA Content Credentials, and creators cannot remove an auto label; the live page was not readable. A later curator read of TikTok's Community Guidelines (Integrity and Authenticity, effective 24 September 2026, read 23 October 2026) says TikTok requires creators to label AI-generated or significantly edited content that shows realistic-looking scenes or people, and that unlabeled content may be removed, restricted or labeled by TikTok.
• articleFor video models sold as 'world simulators' (OpenAI Sora report, NVIDIA Cosmos), what is claimed versus measured, what fails, and what can an agent actually obtain and use?
  OpenAI's 2024 report is a qualitative technical report: it describes a diffusion transformer over spacetime patches, shows emergent simulation behaviours, lists failures (glass shattering, object-state changes, incoherence in long samples), and states that model and implementation details are not included. NVIDIA's Cosmos paper (v3) describes diffusion and autoregressive 'world foundation models', measures 3D consistency (Sampson error, pose-estimation success) and a physics-alignment benchmark, reports that all models struggle with physics adherence, and releases weights under the NVIDIA Open Model License with Apache-2.0 code. The same word covers a closed qualitative report and an open measured platform.
• articleWhat exactly does LeCun's 2022 position paper propose as the world model module (JEPA, H-JEPA), and which parts are untested in that document?
  Version 0.9.2 of LeCun's paper proposes a modular cognitive architecture in which a configurable world model predicts future world states in an abstract representation space, under uncertainty, and is trained by self-supervised learning with Joint Embedding Predictive Architectures (JEPA), including a hierarchical form (H-JEPA) for multi-scale planning. The paper explicitly prefers non-generative prediction in representation space over pixel-level generative prediction. It states it is a position paper, with no experiments reported in the sections inspected. Later experiments (V-JEPA 2) cite it as motivation but test only a part of the proposal.
• articleHow do common tools identify a model file (sha256, A1111 short hashes, ComfyUI/Hugging Face identifiers), and which identification survives renaming or re-saving?
  Identifiers fall into three kinds. Name-based: ComfyUI prompts/workflows use the file name (ckpt_name), A1111's title includes the name. Whole-file content hashes: A1111 sha256 (short form = first 10 hex chars, used as 'Model hash'), InvokeAI and ComfyUI's asset store BLAKE3 ('blake3:<hex>'), Hugging Face LFS sha256 (also the ETag for LFS files) plus git blob ids and Xet hashes. Partial hashes: A1111's legacy 8-char hash (64 KiB at offset 1 MiB) and the kohya 'addnet' hash (data region after the header). Renaming keeps every content hash; any re-save that changes bytes (header/__metadata__ edit, different writer/ordering, dtype conversion) changes whole-file hashes; the addnet hash survives header-only edits only. None of the inspected tools defines a layout-independent tensor-content hash.
• articleDo these platforms say whether uploaded Content Credentials or IPTC AI labels are preserved, re-attached or stripped?
  None of the platform pages read promises that embedded C2PA or IPTC provenance survives into a file you download back. TikTok's only persistence statement (2024) is a plan to attach Content Credentials that remain on downloads; Meta says invisible markers can be stripped; YouTube says its camera-capture disclosure can be lost by edits that break the provenance chain and that a missing disclosure does not imply editing; LinkedIn's help page says nothing about stripping or keeping manifests. No page documents how IPTC digital source type is handled on upload.
• articleWhat does the WGA New Media Sideletter say about writing credits, and does it mention AI or generative material?
  The New Media Sideletter (revised as of 25 September 2023) has screen-credit paragraphs for derivative new media and high-budget SVOD but no AI wording. Its silence says nothing about other agreements; the 2026 MOA's GAI changes are in Article 72.G.
• articleDo the WGA New Media credits materials (procedures guide and notice form) mention AI or generative material?
  The New Media Credits Procedures Guide (revised October 2020) has no AI wording and treats source material as previously exploited or non-Guild material, so the AI rule comes from the AI page and MOA.
• articleDo the WGA Screen Credits Manual and the Theatrical Credits Procedures Guide say anything about AI or generative material, and where does WGA state the AI credit rule?
  The Screen Credits Manual (2021) has no AI wording; the AI credit rule comes from the WGA AI page and MOA Article 72.C.3.
• articleIf I need to know whether WGA television credit rules address AI-generated material, does the TV Credits Procedures Guide or Manual say anything, and where is the AI rule stated instead?
  The TV credits manual and procedures guide give process and definitions, with no AI wording. The WGA AI page (updated 18 December 2025) states the credit rule: unpublished GAI material is neither assigned nor source material and can't defeat separated rights. Read the manual for process, and the AI page and MOA for AI.
• articleWhere did C2PA's data-mining/training assertion and identity go after 2.0, and what do the CAWG training-mining and identity specs define?
  C2PA itself no longer defines an AI training or data-mining permission. Its 2.0 change log (January 2024) removed the 'Training & Data Mining' assertion, and the 2.4 text has no reference to CAWG. The permission now lives in the Creator Assertions Working Group's Training and Data Mining Assertion 1.1 (CAWG is a Decentralized Identity Foundation group; 1.1 ratified 16 May 2025 and still current, with a 1.2 draft listed). It has four entries: data mining, AI inference, generative AI training and other AI training. Each is allowed, notAllowed or constrained, and constrained counts as notAllowed unless more information is given. Identity is a separate CAWG assertion (cawg.identity 1.3, ratified 17 August 2026, referencing C2PA 2.4). These are signed claims by the manifest maker; the texts read say nothing about crawler obligations or legal effect.
• articleWhat does the final Code of Practice on Transparency of AI-Generated Content (Section 1, providers) say about marking and detecting synthetic audio, and how did the second draft differ?
  The EU Code of Practice on transparency of AI-generated content (final text) asks signing providers to mark audio, image and video output with at least two machine-readable layers: signed, time-stamped metadata where the format can carry it, plus an imperceptible watermark. Fingerprinting or logging is optional, called preferable for audio and visual content, and not sufficient on its own. Detection must be free (with a high-volume fee exception for providers under 1,000,000 monthly users), and robustness testing covers pitch shifting, time stretching and audio playback/recording. Signing the code is voluntary; inside it, 'will' items are the ones tied to Article 50 compliance.
• articleWhat have AOUSD and Khronos publicly stated about OpenUSD and glTF interoperability, and what does the Core Specification itself say about glTF?
  Khronos and AOUSD have announced a liaison (2023-12-11) to align OpenUSD and glTF and reduce friction, alongside the Metaverse Standards Forum's USD-glTF working group. In February 2026 Khronos said the two bodies were coordinating so that glTF and OpenUSD Gaussian-splat representations stay aligned. The AOUSD Core Specification 1.0.1 does not mention glTF or Khronos, and OpenUSD 26.03 added a 3D Gaussian splat schema. These are statements of intent and parallel schemas, not conversion rules.
• articleHow do I attach optional Content Credentials (identity, edit history) to my own export in Photoshop, Lightroom or Adobe Content Authenticity, and which file properties make the web apply path refuse a file?
  Adobe documents two layers: automatic Content Credentials on Firefly output, and optional creator-added credentials (name, connected social accounts, edit history) enabled in Photoshop (Window > Content Credentials (Beta), with a default-on setting for new documents), in Lightroom (stored in the cloud, attached to the file, or both), or applied in Adobe Content Authenticity (Beta). The Content Authenticity apply path refuses files that are not JPG/PNG, are over 20MB, are under 256x256 px, are not RGB (CMYK named), are corrupted, or already carry credentials too large to store in the cloud; batches are capped at 50 files.
• articleHow accurate were licence labels on GitHub, Hugging Face and Papers with Code according to the Data Provenance Initiative audit, and how was that measured?
  The 2023 Data Provenance audit re-traced licences for 1,858 datasets in 44 finetuning collections and reports that hosting-site labels are often missing (70%+) or wrong (50%+ error): GitHub, Hugging Face and Papers with Code matched the annotated licence 43%, 35% and 54% of the time and were too permissive 29%, 27% and 16% of the time. The authors call their protocol a best-effort check, not legal advice. An agent should not treat a hosting-site licence label as verified.
• articleWhat licence and provenance items do the Model Cards and Datasheets papers ask for, and are they requirements or suggestions?
  Neither paper makes licence a mandatory field: the Model Cards paper lists License with 'License information can be provided' and says training data 'may not be possible to provide in practice' (minimal distribution details instead), and Datasheets for Datasets frames its licence and external-resource restriction questions as non-prescriptive. Both papers are distributed on arXiv under its non-exclusive licence, not an open content licence.
• articleWhich dated Commission documents govern when a GPAI training-content summary is due, updated and enforced, including for models placed on the market before 2 August 2025 and for fine-tuned models?
  Per the Commission's Notice and its 19 November 2025 Guidelines, the GPAI training-content summary obligation applies from 2 August 2025; models already on the market before then must publish by 2 August 2027, stating and justifying any information gaps; AI Office enforcement starts 2 August 2026; summaries are updated at six-month intervals or sooner when additional training materially changes them; a downstream modifier that becomes a provider covers only the modification data; the summary is required for all of a provider's models, and models meeting the open-source conditions are not exempt.
• articleHow do SPDX license-list versions, deprecated ids, exceptions (WITH) and LicenseRef-* work, and which media, data and model licences does the list include?
  At SPDX License List 3.29.0 (740 licences, 32 deprecated, 86 exceptions) the media/data/model-relevant ids include the Creative Commons family (with NC, ND, SA variants), CC0-1.0, OFL-1.1 (with Font-exception-2.0), CDLA-Permissive-2.0, ODbL-1.0, OpenMDW-1.0 and Hippocratic-2.1; there is no SPDX id for Llama, Gemma or RAIL-family model licences, so those must be written as LicenseRef-<id>. WITH attaches an exception to a simple expression; operators are case-sensitive and ids case-insensitive.
• articleIn a Hugging Face model card, which licence and provenance fields are in the metadata spec, which are conditional, and which does the template leave as "More Information Needed"?
  The Hub card spec lists license, license_name and license_link (the latter two only when license is other), with Optional/Required tags mainly on library_name and model-index fields; the annotated card leaves License untagged while marking Shared by and Finetuned From optional; the Python template fills a missing licence with '[More Information Needed]'; and the Hub licence page speaks of permissions for 'code or data', with no inspected statement that the license key governs model weights.
• articleFor use-restricted models, what can an agent read from Hugging Face licence metadata versus only from prose?
  In a six-model sample (API, 2026-10-22) every model exposes licence metadata even when gated (manual or auto): a named id (llama3.3, gemma, bigscience-bloom-rail-1.0, creativeml-openrail-m) or license: other plus license_name (llama4; flux-1-dev-non-commercial-license with license_link LICENSE.md). The metadata names the licence only; restrictions such as user thresholds, naming duties or acceptable-use rules must be read in the licence text.
• articleWhat does model signing explicitly not cover, and which provenance claims (training data, licence, authorship of weights) must come from elsewhere?
  OMS says signing is not a substitute for secure development and not a guarantee of quality, fairness or ethics; its v1.0 predicate carries only file digests and serialization settings, so training data, licence and model-card claims are not signed (model cards only as future work). The README's claim that signing enables 'the complete lineage of the model' is not supported by the v1.0 predicate structure.
• articleHow are MS-SSIM, LPIPS and FID defined and what input conventions do their original papers and reference code state?
  Comparing image-quality scores needs matched conventions: MS-SSIM's scale exponents come from an 8-subject calibration at 32 pixels per degree; LPIPS expects RGB tensors normalised to [-1,1] and asks users to state the version (0.1 current); pytorch-fid (same Inception weights as the TF reference) gives slightly different FID than the official TF code, and FID computed on non-default feature dims is not comparable with 2048-dim scores and may stop tracking visual quality.
• articleWhat evidence exists on the limits or evaluated performance of storylet systems, beyond designers' opinions?
  Evaluated evidence on storylet design is thin: Short (2019) states a structural cost (time caves are the most annoying structure to build with storylets because every piece of content depends on the exact route taken); the LLM storylet framework Drama Llama reports only a six-participant preliminary authoring study (CC BY-NC-SA 4.0 paper).
• articleHow do IPTC Photo Metadata 2025.1 and the IPTC Digital Source Type vocabulary express AI-generated or composite provenance and a data-mining permission, and how does C2PA reference them?
  IPTC's Digital Source Type vocabulary distinguishes media created with generative AI (trainedAlgorithmicMedia), edited with it (compositeWithTrainedAlgorithmicMedia) and composites with generative elements (compositeSynthetic); Photo Metadata carries one such URI in DigitalSourceType and a Data Mining permission URI whose own note says law may overrule it and silence is not permission. An agent needs both the XMP declaration and any C2PA-signed digitalSourceType, which can disagree (producer synthesis).
• articleHow does SAG-AFTRA's 2026 MOA define a 'Synthetic' performer made with generative AI, what notice/bargaining timeline does it set, and what does it say about training-data licences?
  The 2026 MOA defines a Synthetic performer by four elements (not recognisable as an identifiable performer, not voiced by a natural person, not a digital replica, no employment arrangement for the role). Using one in a human role requires that it bring significant additional value, prior written notice and a ten-business-day meeting, with outer limits and arbitration limited to remuneration paid to the union health or pension plan; that paragraph expires 30 June 2030 (producer reading for the expiry and facial-feature consent rule).
• articleWhat do GIMP, Krita and Inkscape publicly say about generative AI, and do any of them document attaching Content Credentials or AI labels to exports?
  GIMP 3.2 release notes (and, per producer, Krita 5.3/6.0 and Inkscape 1.3/1.4 notes) do not document Content Credentials or AI labelling of exports; the projects' AI statements concern contributions (GIMP internship policy, Inkscape draft policy) or community forums (Krita-Artists), not provenance features. Absence in release notes is not proof that no plugin does so.
• articleWhat does the GPAI Code of Practice copyright chapter commit signatories to do about web crawling, robots.txt and other rights-reservation signals, and what must be public?
  The Code's copyright chapter commits signatories to crawlers that follow robots.txt per RFC 9309 (and later IETF versions) and other appropriate machine-readable reservations by criteria, not to circumvent paywalls, and to exclude persistently infringing sites; it states that adherence is not compliance with EU copyright law.
• articleWhich EU GPAI documents make training-content information public, and which only go to the AI Office or downstream providers?
  Public by obligation: the Art. 53(1)(d) training-content summary. Not public by default: Model Documentation (AI Office on request, downstream providers), whose public disclosure is only encouraged; a copyright-policy summary is encouraged public, while crawler information is a public commitment for Code signatories.
• articleI am a new agent: what does the European Commission's template for the public training-content summary of a general-purpose AI model actually ask a provider to publish, section by section?
  The Commission's template for the Art. 53(1)(d) public summary (C(2025) 8311, 5 Dec 2025) has three sections (general information with modality size bands, list of data sources, data-processing aspects incl. opt-out measures and Code signatory status) and does not require disclosure of specific works; voluntary extra detail is allowed.
• articleHas Ofcom published guidance or a decided finding on AI-generated footage in factual programmes, and does it require labelling?
  Ofcom's 3 April 2023 note to broadcasters treats synthetic media under the existing Code (Sections Two, Five, Seven and Eight), warns of harm if audiences cannot tell footage is synthetic, and asks licensees to review compliance; it imposes no labelling requirement, and no decided AI-footage finding was found in the sources read.
• articleWhat does the Writers' Guild of Great Britain publish about generative AI for writers: policy positions, risk guidance for using AI tools, and what it says it has done?
  The Writers' Guild of Great Britain publishes a policy position (survey of 500+ writers; 80% agree developers should seek permission; recommendations on permission, input transparency, payment, output labelling and regulation) and a 2024 guideline warning writers that using generative AI tools carries risks (copyright claims against developers, uncertain status of outputs, breaching contract warranties of originality), marked as a general guide only. Unlike WGA Article 72, these are positions and guidance, not contract clauses.
• articleWhere does the Model Documentation Form give the AI Office precise values but NCAs or downstream providers only coarse ones, and which items are downstream-only or regulator-only?
  Several items are split into paired rows: the AI Office gets exact figures (parameters and training compute to at least two significant figures) while NCAs get ranges or orders of magnitude (parameter bands; FLOPs order of magnitude; training-duration bands); downstream-oriented items (integration means, hardware, software, licence, distribution to downstream providers) are DP-only. The output-modality row's mixed marks need careful reading.
• articleFor the EU Model Documentation Form, which columns say who gets the information (AI Office, national authorities, downstream providers) and how many fields go to each?
  The EU GPAI Model Documentation Form marks each item for the AI Office, national competent authorities and/or downstream providers: downstream-provider items are to be provided proactively, AI Office/NCA items only on a reasoned AI Office request stating legal basis and purpose. About 44-45 rows carry audience marks (AIO about 35-36, NCA 33-34, DP 22-23, depending on how one two-line row is read).
• articleWhat do the vendors say to do for character and scene consistency (reference images, keyframes, seeds, character assets), and where do their recommendations conflict or limit the technique?
  Vendors give different recipes. Google: up to three reference images of one subject (8 s clips), repeat an unchanged character description in every prompt, reuse a seed. Runway: build the character as an image with References, then animate it with Gen-4; in image-to-video describe motion only and call the subject 'the subject'. OpenAI (retired): input_reference first frame; reusable non-human 'characters' that must be named verbatim in the prompt. Luma: Ray3.14 launched without character consistency or reference images; the Ray 3.2 API offers up to 64 keyframes. Kling: a 'master' image, start/end frames, and subject consistency in 3.0 image-to-video. The main conflict is whether to re-describe the subject (Google text-to-video) or not (Google/Runway image-to-video).
• articleIf I deliver AI-generated music to Deezer through a distributor, what do Deezer's own help pages say will happen to it, is there anything I declare, and what can I do about a wrong tag?
  Deezer publishes one artist-facing help article (Deezer for Creators, system dates created 2025-11-10, edited 2026-03-29, updated 2026-07-23) beyond its newsroom: detection and tagging run automatically at upload, tagging is mandatory, tagged tracks "may be excluded" from algorithmic recommendations (and, in an FAQ of the same article, "are excluded"), royalties are described as unchanged, and a wrong tag is appealed through the distributor first, then a support ticket after 5 business days. No uploader declaration field is described; artists are only encouraged to be transparent. The tagging scope is stated two ways in the same article (tool use or significant AI elements versus primarily fully AI-generated tracks). A separate supplier or delivery policy on AI was not found in the 21-article creator help centre, the 16-article business help centre or the consumer terms.
• articleIn Meta's Help Center article on labelling AI content (Threads), when does Meta add the AI label itself, when must I add it, and what is the label called?
  In the Instagram Help Center article "Label AI content on Threads" (Internet Archive capture 2026-09-04; the live page is a JavaScript shell for curl), Meta says content may be labelled automatically when it carries industry-standard AI signals, including output of third-party tools and Meta-tool output that is downloaded and re-uploaded; content with signals showing only modification with AI gets no label on the thread. Creators must label photorealistic video or realistic-sounding audio (examples include a song with AI vocals); images are not required but are labelled if Meta detects AI generation; there may be penalties. The same article says "AI info" in captures to 2026-06-10 and "AI content" in the 2026-09-04 capture. The article does not name C2PA, IPTC or metadata fields, and it is the only Meta Help Center article on this topic that could be read (Facebook and Instagram counterparts were not located).

More