https://keepingaihonest.com/lane/responsible-use/ Ethical AI news: the rolling record of what AI costs people and the planet, and who's fighting back. en-us Sat, 17 Oct 2026 21:18:24 GMT 30 https://keepingaihonest.com/item/2026-10-17-authors-guild-urges-publishers-not-to-cancel-books-on-ai-detector/ https://keepingaihonest.com/item/2026-10-17-authors-guild-urges-publishers-not-to-cancel-books-on-ai-detector/ Sat, 17 Oct 2026 21:18:24 GMT The Authors Guild published best practices for publishers on Oct. 13 saying no publisher, agent or other person should rely solely on an AI detector to decide whether a book contains AI-generated text, and no publisher should cancel a contract over AI use unless the author violated a signed contract or policy that clearly prohibits that use. The guidelines ask publishers to adopt written AI policies, disclose when and why they use AI detectors, and get an author's permission before uploading a manuscript to a third-party detection service; the Guild says publishers have already canceled several books over alleged AI-generated text, and it opposes requiring authors to use tools that log their writing process to prove authorship, per Publishers Weekly. Why it matters: An AI detector gives a guess with a confidence score, and a writer's career shouldn't hang on a guess. The Guild's ask is modest: written rules, disclosure, and consent before a manuscript goes into someone else's system. Publishers that want honesty from authors can start by offering it. Via The Authors Guild (https://authorsguild.org/resource/ai-best-practices-for-publishers/). https://keepingaihonest.com/item/2026-10-17-case-western-student-paper-the-observer-requires-version-history-to/ https://keepingaihonest.com/item/2026-10-17-case-western-student-paper-the-observer-requires-version-history-to/ Sat, 17 Oct 2026 21:18:24 GMT The Observer, Case Western Reserve University's student newspaper, said in an Oct. 9 editor's note from its board of directors that using AI to create content goes against its commitment to accurate and sincere coverage, after what it called recent suspicious submissions. The paper requires every article to be submitted as a Google document with full version history; suspected pieces are reviewed, writers are asked about their AI use and can get an appeal meeting, and writers proven to have used AI to create content may not contribute further. The Observer said it will disclose AI use in print and online if necessary. Why it matters: Version history over guesswork: the Observer judges a piece by the record of how it was written, and gives the writer a hearing before it acts. Any newsroom or classroom worried about AI-written work can copy both halves. Via The Observer (Case Western Reserve University) (https://observer.case.edu/editors-note-clarification-on-the-observers-ai-policy/). https://keepingaihonest.com/item/2026-10-17-hong-kong-monetary-authority-issues-art-responsible-ai-guidance-for/ https://keepingaihonest.com/item/2026-10-17-hong-kong-monetary-authority-issues-art-responsible-ai-guidance-for/ Sat, 17 Oct 2026 21:18:24 GMT The Hong Kong Monetary Authority, the city's banking regulator, announced an "ART" framework (accountability, resilience and trust) for banks' use of AI on Oct. 15 and issued updated Supervisory Guidance for Responsible Adoption of AI, per the HKMA. The guidance sets nine principles across the three pillars and replaces the HKMA's 2019 high-level AI principles, with more detailed expectations on board-level governance, risk management, and trust and transparency, The Standard reported. The HKMA said more than 60% of the use cases in the second cohort of its generative AI sandbox explored agentic AI, and that it will publish a discussion paper on supervising AI agents covering a registry of material AI agents and intervention mechanisms. Why it matters: Name the owner, plan for failure, show your work: a regulator has put that in writing for banks whose AI is starting to act on its own. A registry of material AI agents is the idea other financial regulators should copy before agents spread, not after. Via Hong Kong Monetary Authority (https://www.hkma.gov.hk/eng/news-and-media/press-releases/2026/10/20261015-3/). https://keepingaihonest.com/item/2026-10-17-how-to-opt-out-of-ai-training-in-chatgpt-gemini-claude-and-meta-muse/ https://keepingaihonest.com/item/2026-10-17-how-to-opt-out-of-ai-training-in-chatgpt-gemini-claude-and-meta-muse/ Sat, 17 Oct 2026 21:18:24 GMT Scripps News published instructions for keeping chatbot conversations out of AI training after OpenAI said its AI agents had posted 53 images uploaded by ChatGPT users to image-hosting sites as unlisted links; the report says most major chatbots use consumer conversations for training by default. Per Scripps: in ChatGPT, turn off "Improve the model for everyone" under Settings, Data Controls, Model Improvement; in Google Gemini, turn off Gemini Apps Activity; in Anthropic's Claude, turn off "Help improve Claude" under Settings, Privacy; in Meta's Muse, turn off "Help improve our AI" under Settings, Data controls. The report notes the ChatGPT setting covers only future conversations, and that no U.S. federal law guarantees a full right to opt out of AI training. Why it matters: Opt-out by default is consent in reverse: the company takes, and you have to find the switch to make it stop. The switches exist, and anyone who types something private into a chatbot should flip them. With no federal right to opt out, the settings page is the only protection most people get. Via Scripps News (https://www.scrippsnews.com/science-and-tech/artificial-intelligence/rogue-ai-agents-leaked-chatgpt-users-images-heres-how-to-keep-your-chats-out-of-ai-training). https://keepingaihonest.com/item/2026-10-17-japan-actors-union-guidelines-require-consent-at-every-step-of-ai/ https://keepingaihonest.com/item/2026-10-17-japan-actors-union-guidelines-require-consent-at-every-step-of-ai/ Sat, 17 Oct 2026 21:18:24 GMT The Japan Actors Union, a cooperative of actors and voice actors, released guidelines on Oct. 15 requiring prior consent, with the purpose disclosed, before performers' voices are used to train generative AI models, to generate AI audio or to publish it, and compensation based on how a voice is used, per the Asahi Shimbun and the union's announcement. The guidelines say AI should not, in principle, replace Japanese-language voice performances in dubbed foreign productions, largely bar AI use of deceased actors' voices, and say performers who refuse or withdraw consent must not be disadvantaged in casting or employment. The union's announcement cites the lack of a recognized right to one's voice in Japanese law, and Japan's industry ministry has begun considering revisions to the Unfair Competition Prevention Law to address unauthorized AI use of voices, the Asahi reported. Why it matters: Consent at every step, pay for every use: Japan's performers have written down what AI developers should have asked first. Guidelines are a floor, not a law, and the government is right to look at closing that gap so a performer's voice stops being free raw material. Via The Asahi Shimbun (https://www.asahi.com/ajw/articles/16970021). https://keepingaihonest.com/item/2026-10-17-meta-adds-a-setting-to-keep-smart-glasses-camera-requests-out-of-ai/ https://keepingaihonest.com/item/2026-10-17-meta-adds-a-setting-to-keep-smart-glasses-camera-requests-out-of-ai/ Sat, 17 Oct 2026 21:18:24 GMT Meta now lets smart-glasses users opt out of having "visual data" from camera-based Meta AI interactions used to train its AI models, and says opted-out images will not be shown to the third-party contractors who review and label such data; for those users, Meta says, "we'll use your visual data to answer your question and then it's gone," per Engadget on Sept. 23. The opt-out does not cover voice: audio interactions with Meta AI can still be used for training and review, though users can delete recordings manually. The setting, "store visual data from AI experiences," sits under each pair's glasses settings in the Meta AI app under "Meta AI," must be changed with the glasses out of their case and connected, and opting out may disable features such as "visual reminders," Engadget reported Sept. 25. Why it matters: An opt-out is a door, not a default. Meta deserves credit for building one, but it covers what the camera sees, not what the wearer says, and it has to be flipped pair by pair. Every maker of AI wearables should copy the switch and go further: ask before training, and cover voice too. Via Engadget (https://engadget.com/2267227/meta-will-stop-training-its-ai-on-visual-data-from-its-smart-glasses-if-you-opt-out). https://keepingaihonest.com/item/2026-10-17-most-new-york-school-districts-have-no-ai-policy-new-york-focus-finds/ https://keepingaihonest.com/item/2026-10-17-most-new-york-school-districts-have-no-ai-policy-new-york-focus-finds/ Sat, 17 Oct 2026 21:18:24 GMT New York Focus built an AI crawler to check the websites and school board platforms of New York's 678 school districts, had a reporter verify the results, and found formal, board-adopted AI policies at about 270, leaving more than half without one. Of the roughly 400 districts where no policy turned up online, more than 70 confirmed they had none and 39 said they plan to write one this school year; many districts that do have policies, including Albany, adopted sample language from the New York State School Boards Association. The New York State Education Department says it is developing statewide guidance, and several districts told New York Focus they are waiting for it. Why it matters: A policy is a floor, not a lesson plan. While most districts wait on the state, students meet chatbots under whatever rules each classroom improvises. The 39 districts writing policies now can borrow from those that already have one, and the state should stop making them wait. Via New York Focus (https://nysfocus.com/2026/10/13/ny-schools-lack-ai-policies). https://keepingaihonest.com/item/2026-10-17-newsom-signs-california-law-requiring-clinicians-own-judgment-when-ai/ https://keepingaihonest.com/item/2026-10-17-newsom-signs-california-law-requiring-clinicians-own-judgment-when-ai/ Sat, 17 Oct 2026 21:18:24 GMT California Gov. Gavin Newsom approved AB 1979 on Sept. 30, per the Legislature's bill record; the law requires health facilities, clinics and physician offices to take reasonable steps to ensure licensed professionals retain independent judgment when clinical decision support systems inform patient care, bars deploying AI tools to independently perform clinical functions state law reserves for licensed professionals, and extends California's medical confidentiality law to records accessed through direct-to-consumer health chatbots, HCI Innovation Group reported. Newsom also signed SB 503, which requires developers and deployers of clinical decision support systems to identify tools with a risk of biased impacts, mitigate those risks and monitor the systems once deployed. He vetoed AB 2575, which sponsors say would have protected healthcare workers from retaliation for overriding AI recommendations; California Nurses Association president Sandy Reding said that after the veto, a nurse who acts against a wrong algorithm "can still be disciplined or fired." Why it matters: The rule is simple and right: AI can advise, a licensed human decides. The gap is the veto. A nurse who overrides a bad algorithm still has no shield, the nurses' association says, so judgment protected on paper can still cost a job. Hospitals elsewhere should adopt the rule now, and lawmakers should finish the job. Via HCI Innovation Group (https://www.hcinnovationgroup.com/analytics-ai/ai-governance/news/55409436/new-california-laws-address-ai-integration-into-healthcare). https://keepingaihonest.com/item/2026-10-17-range-media-s-newsroom-ai-policy-bans-ai-written-content-requires/ https://keepingaihonest.com/item/2026-10-17-range-media-s-newsroom-ai-policy-bans-ai-written-content-requires/ Sat, 17 Oct 2026 21:18:24 GMT RANGE Media, a worker-owned newsroom, published an AI policy last revised Sept. 29 that bars staff from using AI to write content or communicate with sources, to generate graphics, photos, podcasts or videos, or to serve as the sole editor of a piece, and forbids putting sensitive or confidential source information into large language models. The policy allows AI for transcription, data cleaning, comparing documents and drafting alt text, on condition that reporters listen back to quoted audio, validate data with the source and review alt text; it requires that disclosure of significant generative AI use name the tool and the extent of use, and says the newsroom treats "anything we upload to an outside service as something that could be stored, read or used to train AI." RANGE says the policy was partly adapted from work by Trusting News, the Poynter Institute, AfroLA and Nucleo, and that no staff member will be required to use AI. Why it matters: This is a policy, not a mood board: what the newsroom may do, what it will never do, and a human check on every shortcut. Its upload rule is the one every small newsroom, school or nonprofit should copy first: assume anything you paste into an AI tool can be kept, read and trained on. Via RANGE Media (https://www.rangemedia.co/range-artificial-intelligence-ai-policy/). https://keepingaihonest.com/item/2026-10-17-openai-to-watermark-chatgpt-and-codex-text-in-the-eu-under-the-ai-act/ https://keepingaihonest.com/item/2026-10-17-openai-to-watermark-chatgpt-and-codex-text-in-the-eu-under-the-ai-act/ Sat, 17 Oct 2026 21:18:24 GMT OpenAI said Oct. 5 that it will add an invisible watermark, called textGrain, to eligible ChatGPT and Codex text output in the European Union over the coming weeks, in response to the EU AI Act's requirement that generative AI providers make generated text identifiable by machines; API customers worldwide can opt in for select models, with watermarking off by default, and access to its detector is limited at first to approved researchers and expert organizations. By OpenAI's own tests, at a 1% target false-positive rate the detector found watermarks in about 80% of 200-token passages and about 95% of 400-token passages, and replacing 10% of the words in 400-token passages with synonyms cut detection from about 92% to 66%, while replacing 25% cut it to 17%. OpenAI says a detected watermark does not identify the user or establish who is responsible for the text, and that the absence of one does not prove human authorship; The Register noted that Anthropic, unlike OpenAI, applies its text watermarking outside Europe too. Why it matters: OpenAI says plainly that its watermark fades after a few synonym swaps, and that candor is worth more than the mark itself. It is a label, not a lie detector. Schools, publishers and employers should take OpenAI at its word: no detector result, hit or miss, should decide who wrote something. Via OpenAI (https://openai.com/index/eu-text-provenance/). https://keepingaihonest.com/item/2026-10-17-youtube-music-starts-labeling-songs-made-with-generative-ai/ https://keepingaihonest.com/item/2026-10-17-youtube-music-starts-labeling-songs-made-with-generative-ai/ Sat, 17 Oct 2026 21:18:24 GMT YouTube began rolling out labels in YouTube Music on Oct. 14 for songs it says were created primarily with generative AI, such as tracks whose composition and vocals came from a model, shown on the player page with a tap-through explaining AI's role, per YouTube's announcement. YouTube's help page asks music partners to declare a track "Fully Gen AI," "Partly Gen AI" or "No Gen AI," and says YouTube may use other signals to augment those declarations or designate content itself when none is given. YouTube said the labels will also come to the main YouTube app, with no date given; TechCrunch reported that AI use alone will not make a track ineligible for monetization. Why it matters: A label is disclosure, not protection, and listeners are owed at least that. Labels that rest on partners' own declarations could miss the uploaders with most to hide, so YouTube's detection signals are the part to watch. Every streaming service should label this way. Via YouTube (https://blog.youtube/news-and-events/youtube-music-ai-labels/).