AI Legal & Governance for Membership Bodies? UAE sets the pace

All Membership Bodies have a library of legal documents, policy documents and many have standards as part of the governance process and charitable status. Many organisations diligently help shape policy change through research and influence. These collected data points drive change through drafting and curating new and old policies. It is an expensive, patient and often thankless task. Could this now be revolutionised by AI saving vast amounts of time for the human reviewers?

UAE invests in AI to draft the laws of the future

For those who have missed this over the past few days, the United Arab Emirates is stepping up from proof of concepts to becoming the first country in the world to bring artificial intelligence directly into the lawmaking process.

In an announcement made last week, Sheikh Mohammed bin Rashid Al Maktoum, Vice President of the UAE and ruler of Dubai, shared that the country is launching a new initiative: an AI-powered system designed not just to streamline tasks, but to actively shape and refine legislation. This means AI will be involved in drafting new laws, reviewing existing ones, and even suggesting updates based on how laws play out in real life.

The initiative will be led by the UAE’s Regulatory Intelligence Office, which will oversee this “AI-driven legislative system.” The technology will draw on data from court rulings, government procedures, and the lived effects of legislation. The idea will be to help lawmakers make more informed, responsive decisions, and do so faster.

“This new legislative system, powered by artificial intelligence, will change how we create laws, making the process faster and more precise,” Sheikh Mohammed said on X, calling it a “paradigm shift” for government.

https://thehill.com/policy/technology/5264179-united-arab-emirates-artificial-intelligence-laws/

Elsewhere, Google ran a recent pilot scheme in the UK that showed administrative workers could save over 120 hours a year using AI in tasks. Debbie Weinstein, Google’s Europe, Middle East and Africa president, said, “People wanted ‘permission to prompt'”, in an interview. “‘Is it okay for me to be doing this?’ And so giving them that reassurance was really important.”

The two news stories have connected threads.

Standards and Policy Development – London

Anyone working within a Membership Body that is tasked with keeping up to date with changing policy, regulations and legislation knows how time consuming it can be. Keeping abreast of shifting information, and then interruption for context within the organisation, is complex and often involves several committees.

This week, back in London, there was a small group of people discussing a less ambitious and more cost effective approach to maintaining their environmental Standards, policies and best practices using AI within a domain-specific application. These type of services allow for large amounts of data and documents to be simply ingested and analysed using AI against organisational defined prompt models. The many parallels with the UAE’s and Google’s project were apparent.

Q: “Can we benchmark ours and create a gap analysis across a range of similar policies? This takes us months manually.”

‘Yes something AI has been doing well for a few years.’

Q: “Can we get a clause by clause improvement recommendation based on current legislation or regulative requirements. This takes forever!”

‘Yes something AI has been doing well for a few years.’

Q: “Can we tell the AI our style guide and the outputs use our organisational language and document structure?”

‘Yes something that can be included in the prompt.’

Q: “Can we create staff training materials from the updates?”

‘Yes something that AI has been doing well for a few years. It can also recommend situational learning applications of the clause sentiment, helping your team understand why it is important and when it applies.’

Q: “Can our team still edit the outputs before release?”

‘Yes, the human feedback loop is important. A good AI domain-specific service has an editorial feature that allows for Peer review of outputs and collaboration. Once done, you can then publish privately or push to public via the service.’

Q: “What are you doing different to the UAE?”

‘We don’t charge c.$1.6bn for the service!’

Trying out AI against core policies and guidelines can feel like a big leap, but for membership-led organisations, proof of concepts (PoCs) offer a safe, manageable way to start. By testing AI tools on a smaller scale, teams can see how well the tech works with their data. What admin tasks can be identified as time consuming? Are there tasks you must run every week, month or quarter? These early trials are a great way to uncover any data gaps, explore what’s possible, and build trust in AI. For organisations keen to reduce the burden on their busy teams and make better decisions, PoCs are a smart first step into the AI world.