Leahy's concerns echo those of other AI experts, including AI pioneer Geoffrey Hinton, who recently left Google to warn about the existential risk posed by digital intelligence. The summit, scheduled for 1-2 November, aims to provide a platform for heads of state and tech executives to discuss these issues. However, Leahy argues that vague promises for responsible AI development will not suffice and could result in a victory for tech companies over regulators.
Key takeaways:
- Connor Leahy, CEO of AI safety research company Conjecture, has warned that the upcoming international AI safety summit risks achieving little due to powerful tech companies attempting to influence the meeting.
- Leahy and others believe there should be a complete halt on developing artificial general intelligence, AI models that can perform tasks at a human or beyond-human level of intelligence.
- Leahy and others have launched a campaign called Control AI to urge policymakers to implement a pause on developing the most advanced forms of AI at the Bletchley Park summit.
- AI pioneer Geoffrey Hinton recently quit Google to warn about the 'existential risk' posed by digital intelligence, a sentiment echoed by Leahy and other senior executives in the AI industry.