The UK and US have entered right into a Memorandum of Understanding (MOU) on AI security.
US Trade Secretary Gina Raimondo and British Technology Minister Michelle Donelan expressed shock on the bilateral agreement. Minister Donelan described AI as “the defining technological challenge of our generation”.
The partnership builds on commitments made in the course of the yr AI Security Summit at Bletchley Park in November 2023.
This groundbreaking summit brought together AI industry leaders and political representatives from multiple nations, including a rare collaboration between the US and China.
The summit also led to the creation of AI Safety Institutes within the UK and US dedicated to evaluating open and closed source AI systems.
Minister Raimondo said of the agreement: “It will speed up the work of our two institutes across the total spectrum of risks, be it to our national security or to our society as a complete.”
Raimondo continued, “Our partnership makes it clear that we usually are not running from these concerns – we’re running toward them.”
With this latest agreement, UK and US researchers will conduct joint security assessments, conduct joint test exercises, engage in red teaming and share expertise.
Expressing confidence in making a secure path forward for AI, Donselan said: “Only by working together can we tackle the risks of technology head-on and harness its enormous potential to assist us all live simpler, healthier lives , emphasizing the worldwide implications of the role of the audience in AI security.”
Regulation of the AI industry is stitched together through a patchwork of voluntary agreements and frameworks.
Last yr, for instance, the Biden administration expanded its offerings voluntary security framework to technology firms equivalent to Adobe, IBM, Nvidia and Salesforce, joining existing participants equivalent to Google, Microsoft and OpenAI.
Such voluntary commitments are increasing, but many doubt their effectiveness. Talk is affordable and the tech industry has traditionally not been successful at regulating itself.
AI regulation continues to be largely lacking within the UK
This bilateral agreement might be crucial for the United Kingdom, which lacks almost any type of AI regulation as a consequence of the absence of the European Union's AI law.
The AI law, which begins to be phased on this yr, requires transparency and risk assessment for AI systems. The UK is not going to robotically comply with EU rules after Brexit and has been slow to set its own rules.
Prime Minister Rishi Sunak wanted to advertise a “pro-innovation” framework within the UK, hinting at a deregulated environment.
This US-UK agreement distinguishes the UK from the EU regulatory environment. Sunak and Chancellor Jeremy Hunt had previously discussed constructing a US-inspired tech industry – a British equivalent of Silicon Valley.
However, the UK has up to now did not launch a generative AI startup comparable to any within the US, France's Mistral or Germany's Aleph Alpha.
