UK and US announce a strategic partnership on AI safety and governance

TL;DR:

  • UK and US announce a significant partnership for AI safety and governance.
  • The collaboration merges the White House AI executive order with the UK’s Frontier AI Taskforce.
  • US unveils the “United States AI Safety Institute,” focusing on comprehensive AI risk mitigation.
  • UK established an “AI Safety Institute” to evaluate emerging AI technologies.
  • Both nations commit to information sharing, research collaboration, and outreach to global safety institutes.
  • White House executive order mandates AI developers to share safety test results with the US government.
  • UK’s established Frontier AI Taskforce expedites its contribution.
  • The shared goal is to ensure the safety of AI models before their public release.

Main AI News:

In a groundbreaking move poised to reshape the landscape of artificial intelligence (AI) safety, the United Kingdom and the United States are set to unveil a strategic partnership. This significant collaboration, set to be officially announced, is the result of a close working relationship between London and Washington.

This pioneering initiative seeks to marry the recent AI guardrails established by the White House through an executive order (EO) with the ongoing efforts of the United Kingdom’s “Frontier AI Taskforce.” The synergy between these two AI powerhouses promises to set new global standards in AI safety and governance.

A senior UK official, closely involved in the planning of Britain’s AI safety summit, revealed, “We plan to announce a close bilateral collaboration with the US safety institute this week.” The official, granted anonymity to speak candidly about the summit, disclosed that the event would take place at the historic Bletchley Park on November 1 and 2.

The upcoming summit will witness both nations introducing their own versions of AI safety institutes, with US Vice President Kamala Harris leading the way. During a speech in London, Harris will unveil the “United States AI Safety Institute,” which will be headquartered at the Department of Commerce. According to a US official, who requested anonymity, this institute will focus on crafting guidelines, standards, and best practices to assess and mitigate a comprehensive range of AI risks.

We must address the full spectrum of risk, from potentially catastrophic risks to societal harms such as bias, discrimination, and the proliferation of misinformation,” added the US official, highlighting the comprehensive approach of the new institute.

Simultaneously, British Prime Minister Rishi Sunak has committed to establishing an “AI Safety Institute” tasked with the examination, evaluation, and testing of emerging AI technologies. Sunak emphasized that this institute would build upon the groundwork laid by the Frontier AI Taskforce, which has already gained “privileged access” to technology models from leading AI companies, including Google DeepMind, Anthropic, and OpenAI.

In addition to their respective institutes, both countries are poised to engage in information sharing and research collaboration. Furthermore, the United States intends to extend its hand in cooperation with similar safety institutes in other countries, solidifying its commitment to global AI safety.

The White House’s recently signed executive order mandates that developers of the most powerful AI systems share safety test results and crucial information with the US government before making them public. This regulatory framework will culminate in the establishment of an “AI Safety and Security Board” to oversee and ensure the safety of AI systems.

Ben Buchanan, the White House special adviser on AI, emphasized the substantive nature of this endeavor, stating, “We’re trying to lead with substance here and engage with other countries with substance. This is a vision for how the United States is approaching AI policy and governance.”

Michelle Donelan, the UK’s Tech Secretary, highlighted the expeditious nature of the collaboration, explaining that the UK’s existing infrastructure, particularly the Frontier AI Taskforce, allows it to move swiftly in this arena. She remarked, “We can do it a lot quicker because we already have that initial organization in the Frontier AI Taskforce. We’ve got the expertise, funding, and processes in place.”

Conclusion:

The UK and US partnership in AI safety represents a transformative step towards establishing global standards for responsible AI development and governance. By combining their expertise, regulatory frameworks, and commitment to comprehensive risk assessment, these two AI powerhouses are poised to shape the future of AI safety. This collaboration will not only ensure the responsible deployment of AI technologies but also foster international cooperation in addressing AI-related challenges, promising a more secure and ethical AI landscape for businesses and consumers alike.

Source