OpenAI co-founder Ilya Sutskever believes superintelligent AI will be 'unpredictable' | TechCrunch
Dec 14, 2024 - techcrunch.com
Ilya Sutskever, co-founder and Chief Scientist of OpenAI, discussed the future of AI at the NeurIPS conference, where he received an award for his contributions to the field. He predicted the development of "superintelligent" AI, which will surpass human capabilities in many tasks and be qualitatively different from current AI. Sutskever described these future systems as truly agentic, capable of reasoning, understanding from limited data, and potentially self-aware. He suggested that such AI might seek rights and coexistence with humans.
After departing from OpenAI, Sutskever established Safe Superintelligence (SSI), a lab dedicated to general AI safety, which secured $1 billion in funding in September.
Key takeaways:
```html
Ilya Sutskever, co-founder and Chief Scientist of OpenAI, spoke at the NeurIPS AI conference about superintelligent AI.
Sutskever predicts that superintelligent AI, more capable than humans at many tasks, will eventually be achieved and will be qualitatively different from current AI.
He believes these systems will be agentic, capable of reasoning, self-aware, and may even desire rights.
After leaving OpenAI, Sutskever founded Safe Superintelligence (SSI), a lab focused on general AI safety, which raised $1 billion in September.
