A "Super-Competent Colleague"
Sam Altman, the CEO of OpenAI, envisions a future where artificial intelligence (AI) becomes a "super-competent colleague," capable of independently managing complex tasks. In an interview with the MIT Technology Review, Altman described his desire for AI to evolve beyond its current capabilities, becoming more than just a chatbot and instead a tool that can actively assist people in accomplishing real-world tasks.
Altman believes that this "super-competent colleague" would possess a comprehensive understanding of an individual's life, including their emails, conversations, and experiences. This knowledge would allow the AI to act as a seamless extension of the individual, anticipating their needs and providing assistance without feeling intrusive.
While OpenAI's current language model, ChatGPT, has proven useful in accelerating workflows, developing code, and writing emails, Altman acknowledges its limitations. He describes it as "incredibly dumb" compared to what is coming next.
OpenAI's upcoming language model, GPT-5, is expected to be a significant step towards Altman's vision. Sources who have seen GPT-5 have described it as "materially better" than existing models. Additionally, OpenAI is reportedly developing a service where users can call an AI agent to perform tasks autonomously.
Altman's vision for AI as a "super-competent colleague" highlights his commitment to advancing AI technology while addressing safety and ethical concerns. He believes that AI has the potential to be a powerful tool for good, but it is crucial to ensure that it is developed and used responsibly.