The United Kingdom (U.K.) recently hosted the first Artificial Intelligence (AI) Safety Summit. It is taking place at Bletchley Park, a renowned location that served as the home to codebreakers during the Second World War.
The Main resolution was to establish shared understanding of frontier AI opportunities and risks. “Frontier AI” refers to powerful generative AI models with potential risks to public safety.
28 countries including the US, China, Japan, UK, France, and India, plus the EU, agreed on a declaration addressing AI risks.
South Korea will host a virtual AI summit in 6 months, France will host the next in-person summit in a year.

Bletchley Park Declaration

A “world-first” accord to address AI safety concerns.
Focuses especially on advanced AI models from major entities like OpenAI, Google, and Microsoft.
Recognizes risks from misuse or control issues, especially in cybersecurity, biotech, and disinformation.
Emphasizes international cooperation to address risks.

About AI

· A branch of computer science creating machines that mimic human intelligence.

· Further, enables machines to model or surpass human mind capabilities.

· Examples: self-driving cars, generative AI tools like ChatGPT, and Google’s Bard.

Need for Regulation

· AI tools and models can be opaque, even to experts.

· AI decisions can have inaccuracies, embedded biases, and produce discriminatory outcomes.

· Rise in AI-generated media like images, voice alterations, and deep fakes affecting political and social arenas.

· AI systems collect personal data, raising privacy concerns.

· Concerns about AI becoming uncontrollable or acting maliciously.

US Actions on AI

US President Joe Biden issued an executive order on AI safety, focusing on overseeing AI benchmarks.
Must for AI companies to share test results of new products with the federal government.
Rules to introduce watermarks on AI-enabled products and standards for biotech firms using AI.
Federal agencies to implement AI changes, setting industry best practices.

Global AI Regulatory Landscape

European Union (EU) proposes an AI Act based on AI risk and invasiveness.
UK adopts a light-touch approach to foster AI innovation.
US takes a moderate stance, with Biden’s order laying groundwork for AI regulation.
China has its own AI regulations.
However, notable tech leaders, including Elon Musk and Steve Wozniak, called for a six-month pause in AI development due to uncontrollable AI advancements.

India’s Stance on AI

Union Minister of state for IT, Rajeev Chandrasekhar emphasized AI safety, trust, and avoiding weaponization via social media.
Further, Prime Minister Narendra Modi called for ethical AI tools and a global framework.
Need to move towards a risk-based, user-harm regulatory approach.
Proposals are made to establish a domestic authority for AI and to differentiate between online intermediaries.

Challenges

Concerns about AI include privacy, system bias, and intellectual property rights violations.
Recently, OpenAI founder Sam Altman called for an international AI regulatory body similar to nuclear non-proliferation oversight.

Bletchley Declaration on AI