US and UK sign agreement on AI security testing

Utkarsh Classes Last Updated 06-04-2024

US and UK sign agreement on AI security testing

Agreements and MoU 5 min read

The United States and the United Kingdom have signed an agreement to work together on testing advanced artificial intelligence (AI) models on April 1, 2024.

The agreement follows commitments made at the Bletchley Park AI Security Summit last year.

Officials from both countries signed off, following commitments announced at the AI Security Summit at Bletchley Park in November.
On this, Commerce Secretary Gina Raimondo from the US side and British Technology Secretary Michelle Donelan signed a memorandum of understanding in Washington to jointly develop advanced AI model testing.
The agreement comes at a time when the world is trying to establish safeguards around the rapid proliferation of AI systems.
AI offers many opportunities but also poses significant threats to various social systems, from misinformation to election integrity.

Main points of the agreement between USA and UK:

Under AI partnership, both the countries have agreed on many issues related to it, which are as follows:
- Sharing important information about the capabilities and risks associated with AI models and systems
- Sharing fundamental technical research on AI safety and security fosters collaboration and innovation
- They will work on aligning their approach towards safely deploying AI systems
- And will work to align their scientific viewpoints
- They will work together to accelerate and rapidly iterate a robust suite of assessments for AI models, systems and agents
- The US and UK AI security institutes will also create a common approach to AI security testing
Plans have been drawn up to share our capabilities to ensure that these risks can be addressed effectively.

UK and USA setting up institutes for AI:

Britain said in October that its institute would investigate and test new types of AI. The US said in November that the USA was launching its own security institute to evaluate the risks from marginal AI models. The USA now works with 200 companies and institutions.
The UK and USA plan to conduct at least one joint test exercise on a publicly accessible model. Are considering exploring exchange of personnel between institutions. Both entities are actively seeking partnerships with other nations to advance AI security efforts.

About Artificial Intelligence (AI):

AI began in the 1950s. Artificial intelligence means intellectual ability developed artificially.
Computer system or robotic systems are created through AI. An attempt is made to run it on the basis of the same logic on the basis of which the human brain works.
According to John McCarthy, the father of AI, it is the science and engineering of creating intelligent machines, especially intelligent computer programs. That is, it is the intelligence displayed by machines.
AI is a way of creating computer-controlled robots or software that thinks with intelligence like humans.
It studies about AI on various dimensions. How the human brain thinks and learns, makes decisions and works when solving problems.