Abstract
Intentional and accidental harms arising from the use of AI have impacted the health, safety and rights of individuals. While regulatory frameworks are being developed, there remains a lack of consensus on methods necessary to deliver safe AI. The potential for explainable AI (XAI) to contribute to the effectiveness of the regulation of AI is being increasingly examined. Regulation must include methods to ensure compliance on an ongoing basis, though there is an absence of practical proposals on how to achieve this. For XAI to be successfully incorporated into a regulatory system, the individuals who are engaged in interpreting/explaining the model to stakeholders should be sufficiently qualified for the role. Statutory professionals are prevalent in domains in which harm can be done to the health, safety and rights of individuals. The most obvious examples are doctors, engineers and lawyers. Those professionals are required to exercise skill and judgement and to defend their decision making process in the event of harm occurring. We propose that a statutory profession framework be introduced as a necessary part of the AI regulatory framework for compliance and monitoring purposes. We will refer to this new statutory professional as an AI Architect (AIA). This AIA would be responsible to ensure the risk of harm is minimised and accountable in the event that harms occur. The AIA would also be relied on to provide appropriate interpretations/explanations of XAI models to stakeholders. Further, in order to satisfy themselves that the models have been developed in a satisfactory manner, the AIA would require models to have appropriate transparency. Therefore it is likely that the introduction of an AIA system would lead to an increase in the use of XAI to enable AIA to discharge their professional obligations.
The Accountability Gap in AI
Current AI regulation primarily targets corporations, which are legally structured to shield individuals from personal liability. When an AI system causes harm—such as a biased loan algorithm or a medical diagnostic error—it is often difficult to hold any specific human accountable. Researchers at University College Dublin identify a "dearth of practical proposals" on how to move from high-level ethics to ground-level enforcement. To address this, they propose the creation of a new statutory profession: the AI Architect (AIA).
Methodology: Learning from Established Professions
The researchers used a multi-disciplinary approach to examine how "statutory professions" (medicine, law, engineering) function. They surveyed the European Database of Regulated Professions, specifically filtering for 229 regulated professions in Ireland. By identifying roles that require Professional Indemnity (PI) insurance, they mapped out the existing landscape of high-risk professional conduct.
This mapping revealed that regulated professions requiring PI insurance fall into three clear buckets: Health, Rights, and Safety. These align perfectly with the areas most impacted by AI harms, providing a precedent for the AIA role.
The AI Architect (AIA) Framework
Much like a doctor or a civil engineer, the "AI Architect" would be a legally protected title. Only a registered AIA would have the authority to sign off on high-risk AI solutions. By requiring an AIA to carry professional indemnity insurance and face potential "striking off" from a register for misconduct, the framework introduces personal stakes that corporate structures currently lack.
The AIA would oversee the entire lifecycle of an AI system, ensuring that "human-in-the-loop" is a legal reality rather than a buzzword. This "farm to fork" responsibility spans from initial data preparation to post-deployment monitoring.
XAI as a Professional Necessity
A central insight of the proposal is that if an AIA is legally liable for a model's behavior, they will demand transparency. Explainable AI (XAI) would evolve from a "nice-to-have" feature into a mandatory tool for the AIA to discharge their professional obligations. The AIA would be responsible for providing certified explanations of decisions within the regulatory framework, ensuring that models are not just accurate, but justifiable.
Regulatory Alignment and Implementation
The authors highlight that the EU AI Act (specifically Articles 14 and 29) already requires human oversight by "competent, properly qualified" persons. The AIA is the logical fulfillment of this legal requirement. To overcome potential corporate resistance to new overhead, the authors suggest a phased approach, starting with a requirement for AIAs in all government procurement of AI systems.
Conclusion and Future Outlook
While the introduction of a statutory profession may introduce economic friction, it is presented as a necessary "cost of doing business" in high-risk domains, similar to the pharmaceutical industry. Future work must focus on defining the specific educational standards and global coordination required to make the AI Architect a cornerstone of safe and accountable AI governance.
Summary updated