Speaking at his alma mater, Dartmouth EngineeringMurati highlighted the rapid progress of AI Capabilities.“If you look at the trajectory of improvement, systems like GPT-3 were perhaps toddler-level intelligence,” she explained. “And then, systems like GPT-4 are more like a high school student’s smart intelligence. And then, in the next couple of years, we’ll look to doctoral intelligence for specific tasks.”
Next-gen ChatGPT will ‘affect everything’
This advancement in AI capabilities could have far-reaching implications across various industries. Murati believes that the impact of AI will “affect everything” in terms of cognitive work and work. She noted that while some jobs could be affected, particularly in creative fields, AI could also democratize creativity by lowering barriers.
“It’s much easier to direct a smarter system by saying, okay, don’t do these things,” Murati said, addressing concerns about AI safety and control. She emphasized that intelligence and safety go hand in hand, comparing it to training a more intelligent dog versus a less intelligent dog.
THE CTO of OpenAI also discussed the potential of future AI systems to connect to the internet, interact with each other, and collaborate seamlessly with humans. However, she stressed the importance of putting security measures in place alongside technological advancements.
“In terms of safety, security and the societal impact of this work, I think those things are not an afterthought,” Murati said. “You kind of have to build them alongside technology and in a deeply integrated way to do it right.”
While news of more advanced AI may excite some, it also raises questions about potential risks and societal impacts. Murati acknowledged that “it is not possible to have zero risk” with AI technology, calling for shared responsibility between developers, users, civil society and governments.
AI everywhere: transforming our world and empowering humanity
