In a notable diplomatic breakthrough, countries including the United States, European Union, China, and the United Kingdom have signed a communiqué outlining the risks associated with advanced artificial intelligence (AI) technologies. The move comes as Vice President Kamala Harris delivered a speech at the AI summit hosted by UK Chancellor Rishi Sunak at Bletchley Park.

While the UK summit focused primarily on the unrealized existential risks of AI, Harris encouraged the international community to address the "full spectrum" of AI risks, including issues such as bias and discrimination that are already causing harm. The vice president emphasized the urgency of taking action against these threats, which have direct implications for people's daily lives.

Harris's speech also aimed to position the United States as the model for AI regulation, pushing for other countries, including the UK, to adopt a similar approach. One White House official expressed satisfaction, noting that the international community is receptive to the US way of handling AI.

Critics argue that the UK is falling behind other countries, such as the EU, Canada, and now the US, in establishing concrete rules around AI. However, Sunak's lead official, Matt Clifford, disagreed, highlighting the UK's focus on both short-term and long-term AI risks. He stressed that companies are currently training models with considerably higher efficacy than existing ones, representing imminent dangers.

As part of the US's commitment to AI safety, President Joe Biden issued an executive order (EO) on AI. Harris was intentional about making this announcement during her speech, highlighting the US's proactive stance on the matter. This move signals the US's determination to collaborate with the UK, as well as its willingness to establish its own AI Safety Institute alongside Sunak's initiative.

Questions remain about how effectively the US and UK institutes will collaborate, given concerns over national security, particularly as the majority of leading AI companies are based in the United States. Nonetheless, the UK's Frontier AI Taskforce is leveraging its head start by engaging in discussions with prominent AI companies like Google DeepMind, Anthropic, and OpenAI to access their models voluntarily.

Amidst the success of the first day of the summit, Sunak expressed satisfaction with the participation of the US, EU, and China, and regarded the shared communiqué as a sign of progress. Looking ahead, South Korea will host the next AI safety summit in six months, followed by a summit in France a year from now, indicating that other capitals will follow suit in addressing AI regulation.

Overall, Vice President Harris's speech at the AI summit highlighted the need for global collaboration in tackling AI risks. The US's release of the executive order further underscores its commitment to leading the way in AI regulation. As countries move forward, the challenge lies in striking a balance between information sharing and preserving national security in the rapidly evolving AI landscape.