Research preview. Independent expert review is pending. The worked analyses are hypothetical.
Method and corrections Interactive field model
Inspect the system around the AI system. The model is organized by the clauses of the definition: societies, the four objects, the benefits and harmful patterns governance aims at, and the conditions it races against. Apply a cross-cutting test, narrow by lifecycle stage or level, and inspect how definitions, claims, relationships and sources travel together.
Apply a cross-cutting test Whole model Power Legitimacy Distribution Adaptability
Narrow by lifecycle stage or level Lifecycle stage Any stage Research Data collection Model development Evaluation Procurement Deployment Monitoring Incident response Retirement Level Any level Community and organisational use Corporate Sectoral National Regional International Loading interactive controls. Concept links remain available. Field model · whole model
Groups follow the clauses of the definition; they are not a sequence or causal chain. Actual relationships appear in the inspector.
Jump to selected entity Societies …how societies should shape…
actor AI developer or provider actor Deploying organization actor Affected people actor Civil society and independent researchers Incentives …the incentives…
incentive Deployment pressure incentive Competitive race dynamics incentive Liability exposure incentive Procurement leverage
Institutions …institutions…
institution Regulator or public authority institution Internal safety and governance team institution Standards body institution International institution
Infrastructures …infrastructures…
infrastructure Evaluation capacity infrastructure Compute infrastructure Data, language resources, and data labor infrastructure Energy, water, and minerals infrastructure Data work and content moderation Safeguards …and safeguards around AI…
safeguard Risk management process safeguard Impact assessment safeguard Traceability safeguard Audit record safeguard Meaningful human oversight safeguard Supervision and regulatory review safeguard Pre-deployment evaluation and red teaming safeguard Incident reporting and post-deployment monitoring safeguard Complaint and appeal routes safeguard Review and sunset provisions safeguard Compute controls Benefits and harmful patterns …so that its benefits are broadly shared and harmful patterns are addressed…
principle Accountability principle Safety principle Broadly shared benefit pattern Contestable decision pattern Concentration of power pattern Systematically unreliable provenance pattern Over-reliance and deskilling pattern Sycophancy and engagement-driven agreement pattern Homogenisation and loss of authorship Choices and conditions …while they are still choices rather than conditions.
condition Lock-in condition Governance lock-in condition Reversibility window
Infrastructure · Infrastructures established
Data, language resources, and data labor The training and evaluation data a system depends on, the languages and communities it represents or omits, and the human labor that produces and annotates it.
Whose data, whose languages, whose labor, and who is data-poor as a result?
Tests Distribution Power
Lifecycle Data collection Evaluation
Levels National International Recorded relationships Read each statement from its first entity to its second. Status describes our mapping, not certainty of cause.
Data, language resources, and data labor — conditions → Broadly shared benefit
Mapping: asserted
Communities that are data-poor or absent from training data are both more exposed to bias and less able to access benefits.
Inspect evidence · 1 passage Passage summaries are paraphrases, not quotations. Open the source to read the full passage.
Governing AI for Humanity: Final Report (opens in a new tab) Executive summary, para. xli (p. 14)
Beyond misuse and missed use of data, the report identifies missing data: large portions of the globe are data poor. Failure to reflect linguistic and cultural diversity has been linked to bias and may also be a missed opportunity for those communities to access AI’s benefits.
Scope note: An advisory body report representing a majority consensus of its members in their personal capacity. It is not a UN member-state decision. Page locators refer to the PDF's printed page numbers.
Passage checked 2026-09-06 · un-hlab-missing-data
Claims and evidence · 6 source-assertion Confidence: high Contestability: medium
Access to compute, data and talent determines who can participate in AI; the UN advisory body proposes a floor under the AI divide rather than universal access to frontier compute.
Our reading: The 'floor' is a proposal, not an existing arrangement.
Inspect evidence · 1 passage Passage summaries are paraphrases, not quotations. Open the source to read the full passage.
Governing AI for Humanity: Final Report (opens in a new tab) Executive summary, section C 'Common benefits', paras. xxxviii–xl (p. 14)
Effective participation depends on access to talent, compute and data. Compute is described as one of the biggest barriers to entry; the report says it is unrealistic to promise access to compute that even the wealthiest countries and companies struggle to acquire, and instead seeks to put a floor under the AI divide.
Scope note: An advisory body report representing a majority consensus of its members in their personal capacity. It is not a UN member-state decision. Page locators refer to the PDF's printed page numbers.
Passage checked 2026-09-06 · un-hlab-compute-floor
source-assertion Confidence: high Contestability: low
Large parts of the world are data-poor, and failure to reflect linguistic and cultural diversity is linked both to bias and to missed benefits.
Inspect evidence · 1 passage Passage summaries are paraphrases, not quotations. Open the source to read the full passage.
Governing AI for Humanity: Final Report (opens in a new tab) Executive summary, para. xli (p. 14)
Beyond misuse and missed use of data, the report identifies missing data: large portions of the globe are data poor. Failure to reflect linguistic and cultural diversity has been linked to bias and may also be a missed opportunity for those communities to access AI’s benefits.
Scope note: An advisory body report representing a majority consensus of its members in their personal capacity. It is not a UN member-state decision. Page locators refer to the PDF's printed page numbers.
Passage checked 2026-09-06 · un-hlab-missing-data
source-assertion Confidence: high Contestability: low
The African Union treats Africa’s lack of compute, and reliance on AI and data developed elsewhere, as a structural barrier and a risk to sovereignty.
Inspect evidence · 1 passage Passage summaries are paraphrases, not quotations. Open the source to read the full passage.
Continental Artificial Intelligence Strategy (opens in a new tab) Section 1.4.4, p. 28; Executive summary, p. 4
The strategy names the lack of high-performance compute as a main barrier for Africa, notes that all the world’s supercomputers sit in about 30 countries, and that training data usually come from developed countries.
Scope note: Global South regional body. Adopted by the AU Executive Council at its session of 18–19 July 2024 and posted by the AU on 9 August 2024; a strategy, not binding law.
Passage checked 2026-10-09 · au-compute-barrier
source-assertion Confidence: high Contestability: low
India’s AI governance guidelines treat global concentration of compute, data and models as a problem to answer with subsidised access rather than a new AI-specific law.
Inspect evidence · 1 passage Passage summaries are paraphrases, not quotations. Open the source to read the full passage.
India AI Governance Guidelines: Enabling Safe and Trusted AI Innovation (opens in a new tab) Foreword, p. 2; section 2.1, pp. 14–15
The guidelines note that compute, data and models are concentrated among a few global players, and respond by expanding subsidised access to compute and data rather than with a new AI law.
Scope note: Non-binding guidelines; they state that a separate AI law is not needed at present.
Passage checked 2026-10-09 · india-resource-concentration
source-assertion Confidence: high Contestability: low
Large language models tested on a benchmark for 17 African languages performed significantly worse than on high-resource languages such as English.
Our reading: One benchmark of 2024 models on African languages; gaps for other languages and later models may differ, but the direction matches other evaluations.
Inspect evidence · 1 passage Passage summaries are paraphrases, not quotations. Open the source to read the full passage.
source-assertion Confidence: medium Contestability: medium
Research with data workers in Latin America found that task instructions impose clients’ views on how data is labelled, and that precarious conditions make workers follow them rather than question them.
Our reading: Qualitative study of specific platforms and one outsourcing firm; it explains a mechanism rather than measuring its prevalence.
Inspect evidence · 1 passage Passage summaries are paraphrases, not quotations. Open the source to read the full passage.
The Data-Production Dispositif (opens in a new tab) Abstract
Analysing 210 instruction documents and 55 interviews, the study finds that task instructions impose clients’ worldviews on data, and that precarious work and economic dependency make workers obedient to those instructions.
Scope note: Qualitative study of data workers in Venezuela and Argentina.
Passage checked 2026-10-18 · miceli-posada-abstract