AI Insights - Is the Legal and Political Landscape Fabourable to AI?

The public sector is also interested in AI and its use. Many governments around the world have developed national strategies to encourage the development and adoption of AI. In 2018, the European Commission presented a plan to invest €20 billion in AI, with a focus on education, research and innovation. In the US, the government launched the American AI Initiative in 2019 to promote AI research and development of related technologies.

The United States led the way in AI-related bills, with a total of 13 between 2016 and 2021. As an example, the 2021 US defence budget included a large increase in spending on civilian AI research and development, to the tune of nearly $2 billion. The Pentagon plans to increase this budget to over $3 billion in 2024. The US government has also begun to implement AI in several of its services to the public, for example to improve wait times for taxpayers when dealing with the Internal Revenue Service or to predict patient health outcomes using Medicare data.

Number of AI-related bills passed, 2016-21 (aggregate)

However, China is still catching up in some major areas (data quality, computing power and research). The Chinese government has put in place several AI-related plans, such as the Next Generation Artificial Intelligence Plan. Chinese institutions have produced 4.5 times more papers than US institutions since 2010 and China is far ahead in areas with security and geopolitical implications, such as surveillance, autonomy and object detection. China's legislation on AI-related risks has been strengthened even though it remains less strict than Western laws for individuals, as evidenced by the adoption of the Data Security Law, the Personal Information Protection Law and the Code of Ethics for Next Generation AI in 2021.

The EU seems to be behind in the AI race, but France is looking to increase its competitiveness. In 2021, the French government has announced a €2 billion bill to invest in new technologies, with almost 50% of it dedicated to skills development and AI training programmes. The French government has developed a strategic plan, with 81 AI laboratories (the largest number among European countries) and 502 AI start-ups.

The UK has only passed five AI-related bills, but has had the most mentions in legislative proceedings in recent years (twice as many as the US). In addition, the UK government introduced its 10-year plan in 2022 to make Britain a global AI superpower. Since 2014, the UK government has invested over ￡2.3 billion in the sector. An investment of ￡250 million has been dedicated to developing the NHS AI Lab at NHSX for the safe adoption of AI in healthcare.

Total mentions of AI in legislative proceedings, 2016-21 (aggregate)

What about government regulations on AI?

In addition to the desire of governments to promote and support the advance of AI in society, they are also seeking to define a legal framework. Existing AI models raise questions about bias, privacy and intellectual property rights. So far, governments have taken various approaches. On the one hand, the UK has opted for a 'light touch' approach to regulation. The US has adopted a similar approach, but the Biden administration is surveying the public to define what a set of rules might look like. The EU is stricter: for example, a law it is proposing, categorises the different uses of artificial intelligence according to their degree of risk. As the level of risk increases - from music recommendations to autonomous cars - stricter monitoring is needed. The UK's The Economist points out that governments "should treat AI like medicine", with a specific regulator, rigorous testing and prior approval before making it available to the public.

The co-founder and current CEO of OpenAI, Sam Altman, even said recently that "a global regulator should be needed for AI", similar to the International Atomic Energy Agency. On the other hand, China has moved in this direction by requiring companies to register their AI products with the government and check their safety before selling them.

We believe that these measures will help to accelerate technological advances, stimulate innovation and strengthen global competitiveness.