8°C

clear sky

TFL Updates
London Daily News

Abu Dhabi puts machine-readable rules into production with OpenAI

Abu Dhabi puts machine-readable rules into production with OpenAI

Abu Dhabi has begun using an AI-supported system to assess routine regulatory compliance in a live food-import pilot, under a new initiative developed with OpenAI. The project converts selected government policies and regulations into machine-readable form so that an AI system can help officials check applications against the relevant rules.

The first deployment is at the Abu Dhabi Agriculture and Food Safety Authority and is focused on food imports. The system is described as already in production, rather than as a future demonstration. Its role is to assist with routine assessments, including checks connected with permits and certificates; the public description does not say that it replaces the authority’s legal requirements or transfers final responsibility away from officials.

Rules remain visible in each assessment

A central design feature is traceability. Each assessment can be linked to the rule that governs it, while government officials retain oversight and accountability for final decisions. That arrangement is intended to let staff see the regulatory basis for an assessment instead of receiving an unexplained output from an automated system.

The initiative, called Machine-Readable Government, is led and governed by Abu Dhabi’s Department of Government Enablement. It takes a defined set of existing policies and regulations and structures them so AI-powered systems can use them when assisting with standard compliance work. The announcement describes support for officials, not a general replacement of human decision-making across government.

For food imports, this approach could help staff apply written requirements consistently when reviewing routine cases. The available details do not specify which individual rules are included, how many applications have been processed, how long reviews currently take, or whether the pilot has changed approval rates. Those measures would be needed to assess its practical effect beyond confirming that the system is in production.

Computing stays in the UAE

Real-time model computation for the pilot runs on graphics processing units in UAE data centres, using OpenAI’s in-country inference infrastructure. This describes where the model computations take place; the announcement does not provide technical performance results or a detailed account of the system’s data-handling controls for this particular use case.

The government’s stated ambition is to make Abu Dhabi an AI-native government. The machine-readable project addresses one part of that goal: preparing selected rules for use by software, so they can be applied as part of an administrative workflow. It does not mean that every law, public service or government decision has been converted or automated.

Part of a wider set of agreements

The launch was announced alongside four agreements between OpenAI and Abu Dhabi-based organisations, witnessed by Crown Prince Sheikh Khaled bin Mohamed bin Zayed after a meeting with OpenAI chief executive Sam Altman. The government initiative is one of the four areas covered.

In energy, ADNOC, G42 subsidiary Inception42 and OpenAI agreed to deploy OpenAI technology across ADNOC operations. A separate arrangement involving Mubadala, Inception42 and OpenAI concerns the use of AI in investment analysis, opportunity identification and risk assessment. Inception42 and OpenAI also agreed to a five-year partnership supporting AI adoption in government, businesses and education in the UAE.

Together, the agreements point to several intended uses, but they do not establish measurable results from those deployments. For the food-import pilot, the confirmed development is narrower and concrete: a defined set of rules is being used in a production system, assessments are traceable to their governing rules, officials retain final accountability, and processing is carried out on GPUs in UAE data centres.

Pin It on Pinterest