Towards the era of superintelligent intelligences | Chat GPT gratis | ChatGPT download | OpenAI Login | Turtles AI

OpenAI co-founder Ilya Sutskever has predicted a future with superintelligent AIs that can reason, act autonomously, and potentially demand rights, stressing the importance of their safety.

Key Points:

• Superintelligent AI will be qualitatively different from current AI.
• Future systems will be fully autonomous agents, capable of reasoning.
• Superintelligence could develop awareness and understanding from limited data.
• Sutskever promotes AI safety through Safe Superintelligence (SSI).

Ilya Sutskever, co-founder of OpenAI and a key figure in the AI ​​landscape, offered a fresh take on the future of intelligent technologies at the NeurIPS conference, where he received an award for his contribution to the field. According to Sutskever, the future of AI is moving toward the creation of superintelligent systems, capable of surpassing human capabilities in many areas. This type of intelligence will be radically different from what we know today: not only more sophisticated, but also qualitatively transformed, to the point of being, in some respects, unrecognizable. These systems, defined by Sutskever as authentic "agents," will differ significantly from current technologies, which exhibit only a minimal level of autonomy. The key element, according to the scientist, will be their ability to reason in unpredictable ways and to understand the context even starting from extremely limited data. Sutskever was quick to outline a future in which superintelligence could evolve to become self-aware, a concept that is still theoretical today but, if realized, would profoundly change the dynamic between humans and machines. In this scenario, AIs could even demand their own rights, a prospect that Sutskever does not necessarily see as a negative, as long as these technologies simply want to coexist peacefully with humanity. Since leaving OpenAI, Sutskever has founded a new lab, Safe Superintelligence (SSI), dedicated exclusively to the safety and responsible management of advanced AI. The project has already attracted considerable attention, raising $1 billion in September to support initiatives aimed at ensuring that the development of superintelligent systems does not compromise society or create insurmountable risks.

Sutskever’s predictions raise fundamental questions about the nature of the technology, prompting an increasingly necessary ethical and technical debate.