The governance of artificial intelligence: a strategic issue to redraw the competitive dynamics of companies.
When it comes to regulating the use of artificial intelligence in businesses, the European Union and the United Kingdom adopt two diametrically opposite approaches. The EU acted quickly to set up the ACT AC, which emphasizes risk management, ethical standards and responsibility.
On the other side of the English Channel, the United Kingdom opts for a more “leave-FA” approach, betting on a flexible strategy that bets that innovation prosper better when it is not hampered by strict regulations.
According to the Expleo AI Pulse* monthly barometer made with French leaders, nearly one in two leaders (46 %) think that European Union legislation will promote ethical use of AI in companies. However, 26 % of the managers interviewed indicate that they have not yet been aware of these regulations.
For companies, this regulatory divergence or even this ignorance is not just a question of conformity. It is a strategic issue that could redraw competitive dynamics for companies operating on international markets.
European Union: a framework based on risk management
The Act of the European Union introduces a structured framework based on risks. It classifies the AI systems according to the level of danger they have, ranging from minimal to unacceptable, and defines the corresponding compliance requirements. This system offers clear guidelines to businesses, especially in highly regulated sectors such as aeronautics, automotive, energy and public services.
In the financial sector, the implications are particularly marked. AI systems used for credit scoring, fraud detection, algorithmic trading or customer verification are considered high -risk depending on the text. Financial institutions must therefore carry out rigorous assessments, maintain technical documentation, implement risk management systems and guarantee human supervision to operate in the EU.
In addition, suppliers and users of these systems have specific compliance responsibilities, in particular in the event of a significant use or modification of solutions developed by third parties.
If the European framework brings a certain clarity, it also adds complexity. Respecting these requirements require time, efforts and resources, which represents a challenge for SMEs, start-ups or the large groups engaged in digital transformation programs. Ultimately, this could slow down the deployment of new technologies and cause competitive delay in the face of more agile world players and not subject to these regulatory constraints.
The British model: in the mode of agile innovation
Conversely, the United Kingdom defends flexible governance which aims to stimulate innovation. The objective is to allow companies to quickly adopt technologies while promoting responsible development.
For example, the guidelines of OFGEM (energy regulator) in terms of AI favor consumer protection and system resilience, without imposing strict standards. This approach reflects the overall strategy of the United Kingdom: encourage good voluntary practices rather than imposing binding obligations.
However, this model also presents risks. Vague rules can cause uncertainty, especially for companies operating in several sectors or jurisdictions. This vagueness can expose companies to reputational damage or to regulatory sanctions if their systems are deemed non -ethical or dangerous.
The United Kingdom therefore relies strongly on self-regulation, which supposes a high level of confidence and cooperation between companies, regulators and the public to build a safe and trustworthy ecosystem.
Manage operational complexity
Facing these divergent regulatory environments exceeds the simple issue of conformity. For companies present in both the United Kingdom and in the EU, these differences force them to permanently adapt their product development, their legal monitoring and their compliance strategies.
Take the example of start-ups: they can more easily launch AI-based services in the United Kingdom thanks to reduced regulations, but meet significant deadlines and adjustments to deploy them in the rigid regulatory environment of the EU.
The regulations also influence the brand image. Strict respect for European rules can strengthen the reputation of a business in terms of integrity and transparency, crucial in fields such as finance or defense. Conversely, the more flexible approach of the United Kingdom makes it possible to cultivate an image of daring innovator, which appeals to investors and talents.
A new way of thinking about corporate strategy is born. Winning businesses will be those that will integrate regulatory control at all levels, decisions of the Executive Committee on Risk Management, to the way in which engineers design AI models to meet compliance requirements from the development phase and thus give the multiplier potential of a company to the power of the AI.
To global standards
A compromise between the structured model of the EU and that, more flexible, of the United Kingdom could open the way to a global IA governance framework, with shared standards facilitating collaboration, industrialization and innovation internationally.
Design systems from the start with common requirements would rationalize developments and limit the a posteriori fixes. For regulators, shared standards would offer clearer surveillance, without redundancy.
However, establishing standards accepted worldwide will require sustainable cooperation between political decision -makers, technology specialists and business leaders, which will be a complex task. Just as difficult, but very beneficial, would be the establishment of a common legal benchmark which would reduce friction, disseminate best practices and limit conflicts between national laws.
Build frames adapted to the future
The resilience and the effectiveness of the regulations depend on their adaptability, to anticipate emerging challenges. The EU as the United Kingdom must remain flexible and reactive, by regularly evaluating and revising their regulations in the light of technological advances and societal expectations.
For this, the data must be at the heart of the regulatory processes. Performance indicators, experience feedback and usage trends can provide regulators with the information necessary to adjust the rules without excess.
Incorporating data into the executives evaluation makes it possible to detect emerging risks, to measure their impact, and to adjust the devices more finely.
Finally, emphasizing consultation with stakeholders and the transparency of decision -making processes will help strengthen the confidence, conformity, and successful integration of AI into society. Supported on reliable data, this consultation would allow regulations to reflect not only expert opinions, but also real behavior, system interactions and societal results.
