Addressing AI’s Future: US and UK Commit to Joint Safety Measures

Key Takeaways

  • The United States and Britain signed a new agreement on AI, following commitments announced at the AI Safety Summit in November 2023.
  • The agreement facilitates technical exchange and collaboration between UK and US AI safety institutes.
  • Through proactive international collaboration, governments aim to tackle concerns over Generative AI and national security risks.

The United States and the United Kingdom have signed an agreement on AI following pledges made at the AI Safety Summit in November 2023.

This partnership establishes the two allies as pioneers in formally collaborating on testing and assessing risks posed by emerging AI models. It addresses mounting concerns surrounding forthcoming next-generation iterations.

U.S. Secretary of Commerce Gina Raimondo and UK Secretary of State for Science, Innovation, and Technology Michelle Donelan signed the document on April 1. It outlines a framework for exchanging technical expertise, knowledge, and skilled personnel in AI safety between the UK’s AI Safety Institute, initiated in November 2023, and its counterpart in the US, which is yet to initiate its activities.

Collaborative Goals

Generative AI, which creates text, images, and video from prompts, elicits excitement and concern. Worries include its potential to render specific jobs redundant, electoral manipulation, and the risk of AI surpassing human capabilities.

Governments recognize the need for proactive safety measures that evolve with technological advancements. The primary objective of the agreement is to jointly advance the testing of sophisticated AI models while mitigating risks associated with national security, potential use in harmful cyber attacks, or the creation of bioweapons.

Furthermore, the institutes will collaborate on the independent evaluation of private AI models from companies like OpenAI and Google. Along with Microsoft and Meta, they are among the technology companies that have pledged to voluntarily open their newest generative AI models for examination by Britain’s AISI at the AI Safety Summit at Bletchley Park.

In conclusion, Britain and the United States seek to establish similar partnerships with other nations to advance AI safety.