SEOUL: At a global summit in Seoul, more than a dozen leading artificial intelligence firms, including OpenAI, Google DeepMind, and Anthropic, renewed their commitments to safety measures, said the British government in a statement.
This agreement, involving 16 tech companies, builds upon the consensus established at the inaugural global AI safety summit at Bletchley Park in Britain last year.
Prime Minister Rishi Sunak emphasized the significance of these commitments, ensuring transparency and accountability in the development of safe AI technologies. As per the agreement, AI firms will disclose their risk assessment frameworks, outlining intolerable risks and mitigation strategies to prevent threshold breaches.
Furthermore, in extreme cases, companies pledged to refrain from developing or deploying models or systems if risks cannot be adequately mitigated. The determination of these thresholds will precede the next AI summit, scheduled to be hosted by France in 2025.
Notable participants in this safety initiative include tech giants Microsoft, Amazon, IBM, and Meta, along with France’s Mistral AI and China’s Zhipu.ai.
Where Integrity is Everything
