Skip to content
NakodaAI

Intelligence / Research & Resources

A research library for AI in the UAE and beyond

Real papers, government strategies, standards, datasets and guides - not summaries dressed up as sources. Every entry links to its own official publication, carries Nakoda's honest read of why it matters, and says plainly how far we checked it.

Catalogued
22
Issuing organisations
22
UAE / MENA relevant
4
Reading paths
5

Reading paths

Curated sequences

Five ways through the library

Not tags - sequences. Each path is a small set of resources Nakoda has put in a specific order, with our own framing for why that order makes sense. Read a whole path, or use it as a syllabus and read at your own pace.

Reading path · For a governance, risk or legal lead building a first AI oversight programme.

Understanding AI Governance

Six documents, in order of how binding they are - not how old they are. Start with values (what "trustworthy" is supposed to mean), move through voluntary frameworks you can actually operationalise, and finish at the one instrument on this list with the force of law. Read in this order and you will see the same handful of ideas - human oversight, transparency, risk-tiering - get progressively more concrete and progressively more enforceable.

  1. Policy documentOECD2019-05

    OECD AI Principles

  2. Policy documentUNESCO2021-11-23

    Recommendation on the Ethics of Artificial Intelligence

  3. GuideEuropean Commission2019-04-08

    Ethics Guidelines for Trustworthy AI

  4. StandardNIST (U.S. Department of Commerce)2023-01-26

    Artificial Intelligence Risk Management Framework (AI RMF 1.0)

  5. StandardInternational Organization for Standardization / IEC2023-12

    ISO/IEC 42001:2023 - Information technology, Artificial intelligence, Management system

  6. Policy documentEuropean Union2024-06-13

    Regulation (EU) 2024/1689 - the Artificial Intelligence Act

Reading path · For a founder, investor or operations lead building an AI-facing business here.

Starting AI in the UAE

Start with why the UAE is doing this at all, then look at what it has actually built. The federal strategy sets the ambition; the Bletchley Declaration is the UAE's international safety commitment; Jais and Falcon are concrete, independently verifiable research output from UAE institutions, not press-release claims. ISO/IEC 42001 and the McKinsey report close the loop - one is the governance credential UAE companies increasingly need, the other is the economic case an investor will ask you to make.

  1. Government reportUAE Office for AI, Digital Economy and Remote Work Applications2017-10

    UAE National Strategy for Artificial Intelligence 2031

  2. Policy documentUK Government (host); signed by 28 states + EU, later 29 + EU2023-11-01

    The Bletchley Declaration by Countries Attending the AI Safety Summit

  3. Research paperInception (G42) / Cerebras Systems / Mohamed bin Zayed University of Artificial Intelligence (MBZUAI)2023-08-30

    Jais and Jais-chat: Arabic-Centric Foundation and Instruction-Tuned Open Generative Large Language Models

  4. Research paperTechnology Innovation Institute (TII), Abu Dhabi2023-11-28

    The Falcon Series of Open Language Models

  5. StandardInternational Organization for Standardization / IEC2023-12

    ISO/IEC 42001:2023 - Information technology, Artificial intelligence, Management system

  6. WhitepaperMcKinsey & Company2023-06-14

    The Economic Potential of Generative AI: The Next Productivity Frontier

Reading path · For an executive who needs the state of the field, the economic case and the risk basics - not the maths.

AI for Business Leaders

Start with the numbers: what the field actually looks like right now, and what generative AI is plausibly worth. Then the workforce report, because "AI will change jobs" is not a plan - this is the employer-reported detail of which jobs and by how much. Close with the two documents that turn all of that into a governance obligation you can actually assign to someone.

  1. Industry reportStanford HAI2025-04

    The 2025 AI Index Report

  2. WhitepaperMcKinsey & Company2023-06-14

    The Economic Potential of Generative AI: The Next Productivity Frontier

  3. Industry reportWorld Economic Forum2025-01

    The Future of Jobs Report 2025

  4. StandardNIST (U.S. Department of Commerce)2023-01-26

    Artificial Intelligence Risk Management Framework (AI RMF 1.0)

  5. GuideEuropean Commission2019-04-08

    Ethics Guidelines for Trustworthy AI

Reading path · For a technical or security lead evaluating what "AI safety" concretely means before it is a policy slide.

AI Safety and Security

Two warnings, bracketing two years of frontier development, then the field's practical answers. Stochastic Parrots raised the alarm on scale and bias before ChatGPT existed; the GPT-4 report shows what a frontier lab actually tested for at release; Constitutional AI is a specific, published training method for making a model harder to misuse without making it useless. The Bletchley Declaration and the NIST framework are where those individual efforts meet actual international and organisational policy.

  1. Research paperUniversity of Washington / Black in AI / Independent2021-03-01

    On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜

  2. Research paperOpenAI2023-03-15

    GPT-4 Technical Report

  3. Research paperAnthropic2022-12-15

    Constitutional AI: Harmlessness from AI Feedback

  4. Policy documentUK Government (host); signed by 28 states + EU, later 29 + EU2023-11-01

    The Bletchley Declaration by Countries Attending the AI Safety Summit

  5. StandardNIST (U.S. Department of Commerce)2023-01-26

    Artificial Intelligence Risk Management Framework (AI RMF 1.0)

Reading path · For anyone who wants the technical lineage, not just the headlines - a computer science student, a new AI hire, or a curious executive.

Foundations of Modern AI

The architecture, the training data, and the map of everything built on top of both. Start with the eight-page paper that changed the field's architecture in 2017, then the two resources - one dataset, one archive - that most large models are trained on some version of. The survey paper is the up-to-date map; the textbook is where to go for the parts the map assumes you already know.

  1. Research paperGoogle Brain / Google Research2017-06-12

    Attention Is All You Need

  2. DatasetStanford University / Princeton University2009-06

    ImageNet: A Large-Scale Hierarchical Image Database

  3. DatasetCommon Crawl Foundation2008

    Common Crawl Web Archive

  4. Academic resourceGaoling School of Artificial Intelligence, Renmin University of China (with Université de Montréal)2023-03-31

    A Survey of Large Language Models

  5. BookPearson2020-04-28

    Artificial Intelligence: A Modern Approach (4th edition)

The full index

Browse

Every catalogued resource

Search by title, author or organisation, or narrow by resource type, topic or geography. This is the complete catalog - nothing here is held back for a separate page.

22 resources

Research paperTechnology Innovation Institute (TII), Abu Dhabi2023-11-28

The Falcon Series of Open Language Models

The technical paper behind Falcon-7B, -40B and -180B - causal decoder-only language models built by Abu Dhabi's Technology Innovation Institute and, at 180B parameters and 3.5 trillion training tokens, among the largest openly documented pretraining runs at time of release.

  • Research Foundations
  • UAE & MENA AI
Verifiedarxiv.org/abs/2311.16867 (opens in a new tab)

Policy documentUK Government (host); signed by 28 states + EU, later 29 + EU2023-11-01

The Bletchley Declaration by Countries Attending the AI Safety Summit

The declaration signed at the UK's AI Safety Summit at Bletchley Park committing signatory states - including the US, China, the EU and the UAE - to international cooperation on frontier-AI safety.

  • AI Governance & Policy
  • AI Safety & Alignment
  • UAE & MENA AI
Verifiedwww.gov.uk/government/publications/ai-safety-summit-2023-the-bletchley-declaration/the-bletchley-declaration-by-countries-attending-the-ai-safety-summit-11-1-november-2023 (opens in a new tab)

Research paperInception (G42) / Cerebras Systems / Mohamed bin Zayed University of Artificial Intelligence (MBZUAI)2023-08-30

Jais and Jais-chat: Arabic-Centric Foundation and Instruction-Tuned Open Generative Large Language Models

The paper introducing Jais, at the time the largest Arabic-centric open large language model, built with a bilingual (Arabic/English) training and vocabulary approach rather than adapting an English-first model after the fact.

  • Research Foundations
  • UAE & MENA AI
Verifiedarxiv.org/abs/2308.16149 (opens in a new tab)

Research paperGoogle Brain / Google Research2017-06-12

Attention Is All You Need

The paper that introduced the Transformer architecture - a sequence model built entirely on attention mechanisms, with no recurrence or convolution - and became the architectural basis for essentially every large language model that followed it.

  • Research Foundations
Verifiedarxiv.org/abs/1706.03762 (opens in a new tab)

WhitepaperMcKinsey & Company2023-06-14

The Economic Potential of Generative AI: The Next Productivity Frontier

McKinsey Global Institute's sizing of generative AI's economic impact across 63 business use cases and four business functions, estimating annual global value potential and productivity uplift through 2040.

  • Economic & Workforce Impact
Verifiedwww.mckinsey.com/capabilities/tech-and-ai/our-insights/the-economic-potential-of-generative-ai-the-next-productivity-frontier (opens in a new tab)

Academic resourceGaoling School of Artificial Intelligence, Renmin University of China (with Université de Montréal)2023-03-31

A Survey of Large Language Models

A continuously updated academic survey covering the background, key findings and mainstream techniques of large language models, organised around pre-training, adaptation tuning, utilisation and capability evaluation.

  • Research Foundations
Verifiedarxiv.org/abs/2303.18223 (opens in a new tab)

Browse by organisation

Publishers

Who is behind this library

  • UAE Office for AI, Digital Economy and Remote Work Applications
  • NIST (U.S. Department of Commerce)
  • International Organization for Standardization / IEC
  • European Union
  • OECD
  • UNESCO
  • UK Government (host); signed by 28 states + EU, later 29 + EU
  • European Commission
  • Government of Singapore (PDPC / IMDA)
  • Stanford HAI
  • World Economic Forum
  • McKinsey & Company
  • Google Brain / Google Research
  • University of Washington / Black in AI / Independent
  • Gaoling School of Artificial Intelligence, Renmin University of China (with Université de Montréal)
  • OpenAI
  • Anthropic
  • Inception (G42) / Cerebras Systems / Mohamed bin Zayed University of Artificial Intelligence (MBZUAI)
  • Technology Innovation Institute (TII), Abu Dhabi
  • Stanford University / Princeton University
  • Common Crawl Foundation
  • Pearson

Elsewhere on Nakoda

Keep going

This library is one part of a larger map

EncyclopediaDefinitions for the terms used throughout this library.Ecosystem directoryThe UAE bodies, regulators and institutions that publish or enforce many of these documents.Nakoda NewsCoverage of what these strategies and standards mean as they develop.AI Governance (services)Where Nakoda helps organisations turn a framework like AI RMF or ISO/IEC 42001 into practice.

How this library works

Editorial method

Nakoda writes none of what is catalogued here

Every entry in this library is a real, independently published paper, government report, standard, dataset, whitepaper, guide, book or policy document. Nakoda redistributes none of it: every official link goes to the publisher’s own page, never a mirror or a paywalled reseller, and no PDF is hosted on this site.

What Nakoda adds is editorial: why a document matters, what its key findings are in plain language, and - honestly, not by default - whether it has a genuine UAE or MENA angle. That analysis is always written in Nakoda’s own voice and never blended into a paraphrase of the source. Each record also carries a verification state, reused from the same scale the Ecosystem directory uses, and says plainly when a link was checked and when a claim is still an open question.

Spot an outdated link or a wrong figure? Tell us
Need help applying one of these frameworks?Explore the divisions