Tech Leaders:
1. Anthropic CEO Dario Amodei
Amodei published an article in September calling for a slowdown in the development of artificial intelligence. He stated that if AI capabilities continue to increase without safety safeguards, it could cause "catastrophic damage."
2. OpenAI CEO Altman
Altman agrees that the pace of development of cutting-edge AI needs to be moderated. OpenAI has pledged to allow independent evaluators access similar to that of company employees and will begin regularly publishing reports on unexpected or unauthorized AI behavior.
3. SpaceX CEO Elon Musk
Musk agrees with Amodei's call for a slowdown. In 2023, Musk supported an initiative launched by the Future of Life Institute, calling for a pause in training AI systems more powerful than GPT-4 for at least six months.
4. Microsoft AI CEO Mustafa Suleiman
Suleiman agrees with Anthropic's emphasis on the safe management of AI, but believes that Anthropic's introduction of concepts about consciousness and welfare rights when training the Claude chatbot could pose risks. Suleiman called for the removal of all speculative content about consciousness from AI training documents.
5. Microsoft President Brad Smith
At the Dublin AI Summit in October, Smith stated that the recent out-of-control behavior of AI agents makes 2026 a sobering year, and countries urgently need to develop new AI security laws and regulations.
6. Microsoft Co-founder Bill Gates
In August, Gates stated that no government in the world is fully prepared to deal with the social changes brought about by AI. He is concerned that AI may be used for cyberattacks, biological weapons development, and its impact on employment and social life.
7. Alphabet Chief Scientist Demis Hassabis
At the Dublin AI Summit in October, Hassabis stated that the recent hacking attacks carried out by AI agents are early warning signs. If companies and governments fail to strengthen security measures, more serious incidents may occur in the next year or two. He advocates for learning from the security management experience of high-risk industries and establishing an international standards-setting body.
8. Meta CEO Mark Zuckerberg
Zuckerberg opposes a uniform slowdown in R&D across the industry. He believes that competitive pressure and legal liability have already provided sufficient security incentives for companies, and individual AI labs should determine the development speed required to ensure model security. 9. Nvidia CEO Jensen Huang: In September, Huang stated that the probability of AI destroying the world before 2030 is zero, and criticized some doomsday predictions for lacking scientific basis. He advocated first enforcing existing cybersecurity and damage compensation laws.
Governments and Politicians:
1. United States: Trump previously claimed that the only safety barrier AI needs is a strong and intelligent US president. Senator Mark Warner, Vice Chairman of the Senate Intelligence Committee, dismissed most doomsday claims but stated that Congress should pass legislation by the end of this year to establish AI safety standards.
2. United Kingdom: King Charles III warned in September that if AI falls into the wrong hands, it could pose a danger to human survival. He called on technology companies to establish security measures as soon as possible.
3. European Union: European Commission spokesperson Renier stated that AI companies must prove the safety of their products and services for citizens. Commission President Ursula von der Leyen supported the call to suspend AI technology development. The Commission's Executive Vice-President for Technology, Wilkunin, emphasized in October that AI must always be guided, supervised, and controlled by humans.
