AI Security Summit 2023: Global Implications and Regulatory Prospects

The Artificial Intelligence (AI) Security Summit held on November 1st in the UK was an event full of significant diplomatic and strategic implications. The meeting, which brought together global leaders, technology experts, academics and civil society representatives, highlighted some key lessons that have a significant impact on the future of AI and its regulation.

The UK has summoned prominent figures, including US Vice-President Kamala Harris and European Commission President Ursula von der Leyen, to Bletchley Park in Milton Keynes, a site of historic significance as a base for decryptors during World War II. All participating governments issued a joint communiqué on the risks and opportunities of technology, affirming the need for an inclusive and people-centred approach to ensure its reliability and security.

Signed by all 28 governments, as well as the European Union, the Bletchley Declaration outlines their shared approach to addressing the risks of any AI model capable of performing a wide range of tasks, developing risk-based policies, and fostering a culture of AI security. The Declaration also commits to intensifying international cooperation in the future. The United Nations confirmed its support for a panel of experts on AI demonstrating a concrete global commitment to address AI challenges.

The United States has demonstrated its power in setting the AI agenda: President Biden has issued a presidential decree requiring technology companies to submit the results of testing powerful AI systems to the government prior to public deployment. In line with the UK’s announcement last week, Kamala Harris announced the creation of an AI Security Institute, effectively strengthening the US position in the field, both politically and commercially.

The event was spiced up by Elon Musk, the richest man in the world and CEO of Tesla, who drew attention to the Summit and widened discussions about AI security to a wider audience. Musk pointed out that AI represents one of the greatest threats to humanity, in addition to having aroused interest in the Summit, highlighting the importance and urgency of the necessary securitisation of AI.

During the Summit, it emerged that the possibility of an existential, but still controversial, risk linked to the threat of an AI that could endanger humanity. However, there is a broader consensus on concerns about short-term risks, such as the spread of disinformation and the malicious use of generative AI in contexts such as elections. Therefore, the various delegations at the Summit claimed the pre-eminence in the regulation of this innovative technology. Although the European Union is close to enacting its AI law, the UK has expressed its support for a less regulated approach due to the rapidly evolving industry. Nevertheless, everyone recognised the importance of international summits such as this one to address AI challenges in a coordinated way.

Ursula von der Leyen, President of the European Commission, proposed a governance framework for AI based on four fundamental pillars: an independent scientific community, setting global standards, reporting procedures and an international alert system. The EU is seeking to take important steps towards real regulation that governs but does not restrict the use of AI, notably with the proposed law providing a risk scale to guide AI applications on four levels, with a particular focus on high-risk systems.

Finally, the 2023 AI Security Summit highlighted the need for global cooperation to address the risks and opportunities associated with AI. The UK has played a leading role in this process, strengthening its diplomatic stance and underlining the importance of setting global standards, developing risk-based policies and fostering a culture of AI security. The international community is recognising the importance of a coordinated approach to address these challenges and ensure that this technology is used responsibly for the benefit of all.

Mondo Internazionale APS - Private Reproduction ® 2023