In a recent address, UK Prime Minister Rishi Sunak discussed the hazards linked to artificial intelligence (AI) technology. Sunak acknowledged the game-changing potential of AI but emphasized the significance of addressing the tangible risks it brings. These risks encompass the potential for AI to be employed in the development of chemical or biological weapons, terrorist activities, cyber attacks, fraud, and child exploitation. Sunak also drew attention to the improbable yet drastic scenario of humanity completely losing control over AI. Nevertheless, he reassured that there is no immediate cause for concern and people should not fret about this matter. The UK government has released an examination of the risks and capabilities of “frontier AI,” which will establish the platform for further conversations at the upcoming AI Safety Summit. The analysis paper identifies numerous risks associated with AI, such as insufficient incentives for developers to mitigate risks, concentration of market power, disruption of labor markets, and the relinquishment of human control over technology. Sunak underscored the importance of the state ensuring AI safety, cautioning against hasty regulations without a comprehensive understanding of AI that would impede innovation. Instead, the UK government is investing in AI safety research and establishing the world’s first AI Safety Institute to assess AI models and investigate possible risks. Sunak aims to achieve a global consensus on AI risks and form an expert panel similar to the Intergovernmental Panel on Climate Change. However, critics argue that solely focusing on “frontier risks” disregards other immediate risks associated with AI, such as transparency, dependability, and accountability. They also express concerns about monopolistic control, competition, and human rights in the AI industry. The forthcoming AI Safety Summit has received criticism for excluding workers and trade unions, despite the widespread implementation of AI in workplaces.
Brought to you by News Live Updates
