Wiki · Whether to govern
Whether to govern — the refusal register
The catalog mostly answers how to govern an AI harm. The methodology §11 limitations admit it “does not equally serve questions about whethera system should be deployed at all.” This register admits that prior question: 11 documented stances that an AI application should be prohibited, paused, banned, or abolished — 5 already enacted in law, the rest pressed by civil society and researchers.
Documentation, not advocacy
Policy Window’s charter §7 commits it to author no advocacy and sign no open letters. This register stays inside that line: it reports each refusal stance neutrally, with its named proponent(s) and a primary source — it does notadopt, endorse, or rank them, and the catalog’s “how to govern” analysis is unchanged. Every source below was adversarially fact-checked (each link verified to resolve and support the attribution); no stance is listed whose source could not be confirmed.
Statutory prohibition
5An enacted legal ban on the practice.- EU AI Act bans AI social scoringAI social scoring of natural persons
Article 5(1)(c) of the EU AI Act prohibits AI systems that evaluate or classify persons over time by social behaviour or personal traits, where the resulting score causes detrimental treatment in unrelated contexts or disproportionate to the conduct.
Proponent: European Union (Regulation (EU) 2024/1689) · Regulation (EU) 2024/1689 (AI Act), Article 5(1)(c) — Prohibited AI Practices (artificialintelligenceact.eu, reproducing the Official Journal text)
- EU AI Act bans untargeted facial-image scrapingUntargeted scraping of facial images to build facial-recognition databases
Article 5(1)(e) of the EU AI Act prohibits placing on the market or using AI systems that create or expand facial-recognition databases through the untargeted scraping of facial images from the internet or CCTV footage.
Proponent: European Union (Regulation (EU) 2024/1689) · Regulation (EU) 2024/1689 (AI Act), Article 5(1)(e) — Prohibited AI Practices (artificialintelligenceact.eu, reproducing the Official Journal text)
- EU AI Act bans workplace and school emotion recognitionAI emotion recognition in the workplace and education
Article 5(1)(f) of the EU AI Act prohibits placing on the market, putting into service, or using AI systems to infer emotions of a natural person in the areas of workplace and education institutions, except where intended for medical or safety reasons. The prohibition became applicable from 2 February 2025 (Article 113(a)); the associated penalty/governance regime applies from 2 August 2025.
Proponent: European Union (Regulation (EU) 2024/1689) · Regulation (EU) 2024/1689 (AI Act), Article 5(1)(f) — Prohibited AI Practices (artificialintelligenceact.eu, reproducing the Official Journal text)
- EU AI Act bans live public facial identification by policeReal-time remote biometric identification in public spaces by law enforcement
Article 5(1)(h) of the EU AI Act prohibits real-time remote biometric identification in publicly accessible spaces for law enforcement, subject to narrow exceptions (e.g. specific serious threats) requiring prior authorisation and safeguards.
Proponent: European Union (Regulation (EU) 2024/1689) · Regulation (EU) 2024/1689 (AI Act), Article 5(1)(h) — Prohibited AI Practices (artificialintelligenceact.eu, reproducing the Official Journal text)
- San Francisco bans government facial recognitionGovernment and police use of facial recognition technology
In May 2019 San Francisco's Board of Supervisors passed the Stop Secret Surveillance ordinance (8-1), the first major US city to bar city agencies, including police, from using facial-recognition technology.
Proponents: San Francisco Board of Supervisors; Supervisor Aaron Peskin · TechCrunch, 'San Francisco passes city government ban on facial recognition tech' (14 May 2019)
Moratorium / pause
1A call to pause, not (yet) a permanent ban.- FLI open letter: pause giant AI experimentsTraining of AI systems more powerful than GPT-4
The Future of Life Institute's March 2023 open letter called on all AI labs to immediately pause for at least six months the training of AI systems more powerful than GPT-4, citing societal-scale risks; it gathered tens of thousands of signatures.
Proponents: Future of Life Institute; Yoshua Bengio; Stuart Russell · Future of Life Institute, 'Pause Giant AI Experiments: An Open Letter' (22 March 2023)
Ban campaign
3An organised campaign for a binding ban.- Stop Killer Robots campaign to ban autonomous weaponsLethal autonomous weapons systems
Stop Killer Robots is a global civil-society coalition campaigning for a new legally binding instrument to prohibit autonomous weapons that select and attack targets without meaningful human control and to ensure human control over the use of force.
Proponent: Stop Killer Robots (coalition of 250+ NGOs) · Stop Killer Robots, 'The story so far' (stopkillerrobots.org)
- Fight for the Future: ban facial recognitionFacial recognition surveillance by government and law enforcement
Fight for the Future's Ban Facial Recognition campaign calls for an outright ban on facial-recognition surveillance by government and law enforcement, arguing the technology's threat to civil liberties outweighs any benefit and that regulation is insufficient.
Proponent: Fight for the Future · Fight for the Future, 'Ban Facial Recognition' campaign (banfacialrecognition.com)
- Stop Scanning Me: reject EU mass message scanningMandatory client-side scanning of private messages (EU 'Chat Control' CSA Regulation)
EDRi and partners' Stop Scanning Me campaign calls on EU lawmakers to reject the CSA Regulation's mandatory scanning of private and encrypted communications ('Chat Control'), arguing mass surveillance and client-side scanning undermine encryption and rights.
Proponents: European Digital Rights (EDRi); Stop Scanning Me coalition · European Digital Rights (EDRi) and partners, 'Stop Scanning Me' campaign (stopscanningme.eu)
Abolition / refusal
2A refusal stance that the system should not exist — dismantle, not reform.- Stop LAPD Spying: abolish predictive policingPredictive and data-driven policing systems
The Stop LAPD Spying Coalition takes an abolitionist stance, organising to dismantle predictive and data-driven policing (e.g. PredPol, Operation LASER) rather than reform them, framing such systems as automating racial violence and banishment.
Proponent: Stop LAPD Spying Coalition · Stop LAPD Spying Coalition, 'Data-Driven Policing' (stoplapdspying.org)
- ACLU v. Clearview AI: court-enforced faceprint sales banClearview AI's faceprint database and mass face-surveillance product
Under Illinois' Biometric Information Privacy Act, the ACLU's lawsuit produced a 2022 settlement permanently banning Clearview AI from selling access to its faceprint database to private entities nationwide, plus added Illinois restrictions.
Proponent: American Civil Liberties Union (ACLU) · ACLU, 'In Big Win, Settlement Ensures Clearview AI Complies With Groundbreaking Illinois Biometric Privacy Law' (2022)
Why this exists
A governance-research engine that only catalogs how to regulate quietly assumes the system will be built and deployed. Surfacing the refusal frame — that some applications should be paused, banned, or abolished outright — keeps that assumption from going unexamined, and serves readers approaching from an abolitionist or refusal-politics standpoint that §11 flagged the catalog under-serves. It complements the accountability measures (does it work, what to do) with the question they presuppose. Machine-readable at /wiki/whether-to-govern.json.