SEOUL: More than a dozen countries and some of the world’s biggest tech firms pledged on Wednesday to cooperate against the potential dangers of artificial intelligence (AI), including its ability to evade human control, as they concluded a global summit in Seoul.
AI safety was the focal point at the two-day gathering. In a recent declaration, over two dozen countries, including the United States and France, agreed to collaborate against threats posed by advanced AI technologies, emphasizing “severe risks.” These threats could encompass AI systems aiding “non-state actors in advancing the development, production, acquisition, or use of chemical or biological weapons,” according to a joint statement from the nations.
The statement also highlighted the risks of AI models potentially “evading human oversight, including through safeguard circumvention, manipulation, and deception, or autonomous replication and adaptation.”
This declaration came on the heels of a commitment made on Tuesday by some of the largest AI companies, including ChatGPT creator OpenAI and Google DeepMind. These companies pledged to share their risk assessments and ensure that they do not deploy systems where risks cannot be managed effectively.
The Seoul summit, co-hosted by South Korea and the United Kingdom, aimed to build on the consensus reached at the inaugural AI safety summit held last year. “As the pace of AI development accelerates, we must match that speed… if we are to grip the risks,” stated UK Technology Secretary Michelle Donelan. She added, “Simultaneously, we must turn our attention to risk mitigation outside these models, ensuring that society as a whole becomes resilient to the risks posed by AI.”
The summit also saw a separate commitment, the “Seoul AI Business Pledge,” from a group of tech companies, including South Korea’s Samsung Electronics and US giant IBM, to develop AI responsibly. Christina Montgomery, IBM’s Chief Privacy and Trust Officer, emphasized the importance of deliberate use of AI. “AI is a tool in the hands of humans. And now is our moment to decide how we’re going to use it as a society, as companies, as governments,” she said. “Anything can be misused, including AI technology. We need to put guardrails in place, we need to put protections in place, we need to think about how we’re going to use it in the future.”
AI’s proponents have hailed it as a breakthrough technology capable of transforming lives and businesses globally, particularly following the immense success of ChatGPT. However, critics, rights activists, and governments have raised concerns about the misuse of AI in various contexts, including election manipulation through AI-generated disinformation like “deepfake” images and videos of politicians.
Many have called for international standards to govern the development and use of AI. Experts at the Seoul summit warned that AI presents a significant challenge to regulators due to its rapid development. “Dealing with AI, I expect to be one of the biggest challenges that governments all across the world will have over the next couple of decades,” remarked Markus Anderljung, head of policy at the UK-based non-profit Centre for the Governance of AI.
Jack Clark, co-founder of the AI startup Anthropic, emphasized that consensus on AI safety cannot be left to tech firms alone, and that government and academic experts must be included in the discussion. “At this summit, I’ve actually been asking every single person I met with: What’s safety to you? And I’ve had a different answer from each person,” Clark told reporters. “And I think that illustrates the problem. You aren’t going to arrive at consensus by the companies alone, and if you did, I doubt it would be the correct one.”
Another key topic at the summit was ensuring that AI is inclusive and accessible to all. It is not just the “runaway AI” of science fiction that concerns experts, but also issues of inequality. Rumman Chowdhury, an AI ethics expert and leader of the non-profit AI auditor Humane Intelligence, highlighted the disparity in AI’s benefits. “All AI is just built, developed and the profits reaped (by) very, very few people and organizations,” she noted.
People in developing countries, such as India, often undertake the labor-intensive tasks necessary for AI’s functionality. “They are the data annotators, they are the content moderators. They’re scrubbing the ground so that everybody else can walk on pristine territory,” Chowdhury added.
- AI Safety: A global commitment to address severe risks associated with AI, including the potential for misuse by non-state actors.
- Corporate Responsibility: Major tech firms pledged to share risk assessments and avoid deploying unmanageable AI systems.
- Inclusive Development: Emphasis on ensuring AI benefits are distributed fairly and not concentrated among a few entities.
- Regulatory Challenges: Recognition of the difficulties in regulating rapidly developing AI technologies.
Key Participants and Commitments
| Participant | Commitment |
|---|---|
| United States & France | Collaborate on AI safety and address severe risks |
| OpenAI & Google DeepMind | Share risk assessments and limit deployment of risky AI systems |
| Samsung Electronics & IBM | Develop AI responsibly under the Seoul AI Business Pledge |
| UK Technology Secretary | Focus on matching AI development speed with risk mitigation |
The Seoul summit underscored the urgent need for international cooperation, corporate responsibility, and inclusive development to harness AI’s potential while mitigating its risks. As AI continues to evolve, stakeholders must work together to establish robust frameworks that ensure the technology benefits all of humanity responsibly and ethically.
Related News:
Featured Image courtesy of DALL-E by ChatGPT