Openai Chairman Brett Taylor puts the bull case for AI agents

We have not yet received a direct definition of Exactly what AI agent is During the Brett Taylor’s mobile world congress in Barcelona on Tuesday. Thehe Sierra The chairman of the founder and the board of Openai preferred that the question of CNN Anna Stewart’s lateral step, asking how “Agentic AI” was “any different chatbot in Genai”, suggesting everyone to hate the first one, but are admired by the “empathy” answers to AI can serve.

Given that his new startup is the construction of a customer service agent, you would expect Taylor to be evangelical in terms of technology potential. And he does not disappoint: “I am more enchanted by large language models and this current wave of technology more than any technology I can remember, maybe since I found the Internet when I was a teenager,” he told conference delegates.

Changing the step with generative AI customer service agents to the earlier iterations of AI Chatbots is just a much higher level of AI capability that can be “multilingual and instant”.

“I think we are now in this era in which these AI solutions are actually better than the alternative,” he said, adding: “We work with companies like Siriusxm in the United States or Adt Home Security, where if your alarm stops working on AI, you will help you to repair it and you don’t have to wait for a team for a team for a team.

“And what is remarkable to these agents is people really like them.”

Superizure Customer experience

These more capable bots for AI services help companies shrink customer service costs that Taylor suggested would help raise the bar as a whole. “I think it will just improve the user experience for so many brands,” he said.

Bots that are very capable can also lead to new challenges, but he acknowledged that he has noted examples in which the AI ​​customer support agent has “hallucinated” recovery policies that do not exist in response to customer compensation.

Therefore, brands developing appropriate “safety rails” for their AI agents are an important piece for safe implementation of the tool, he said. But he was scouring that this challenge would shrink, as customer service agents are becoming more and more in line with the case and policies of each brand.

“Overall, my philosophy is, don’t wait for the technology to be perfect. In fact, this may never be perfect – but the narrow domains you work on, so you can accept these insoluble problems and make them solve, “he said.

“Instead of trying to solve all problems with AI in the world, you narrow it to a domain and say,” Hey, we’ll put some practical security rails around this AI so we can solve problems right now. ” And I think this is an opportunity for every company at this conference, “he said. Along with his own AI customer service company, he names AI code assistant Cursor and supported by Openai legal technology Harvey As examples of AI specialization that successfully applies AI agents in a defined domain.

Taylor is perceived how AI seed agents can become for the brands in the future is also surprisingly maximal. “I think most companies, AI agents will actually be as significant as their website or their mobile application in terms of the percentage of interactions they have with their customers,” he said. “It would not surprise me for most brands here, if in fact, if you quickly progress five or 10 years, their AI agent is their main digital experience, which I think is difficult to imagine at the moment. But I really think the world goes there. “

How people interact with AI agents will probably move away, he also suggested that consumer interfaces interact with these bots will fade more in the background, as technologists are looking for ways to make it even easier to take advantage of technology usefulness.

“I think – I hope – that everyone who looks at their screens all the time will start to melt as a social habit. And with the advent of a conversational AI, when the software can really understand how we say that computers will melt and the devices will melt and I think it will be very exciting, “he said. As a parent, he said, he hopes his own children “not need to stare at the screen throughout his life to engage in technology.”

Redirection

How about the interruption that customer service agents could have in jobs?

Taylor said this was a valid concern, but again he expressed the optimism that the change would eventually be good for humanity – predicting that while some work roles would disappear, new ones would open in their place. But he added that “technology manufacturers are responsible for having this conversation, not just delivering the technology.”

The great risk of changing jobs powered by AI is that the necessary termination will not be able to keep up with the speed of change, he said. “When the interruption happens faster than society can overcome, it is a destructive force. So basically I think this requires a public, private partnership. “

The moderator also asked the chairman of the OPENAI board for AI Giant’s plan to move from a non -profit purpose to a profit venture, which attracted some critical attention.

Taylor said that Openai’s stated mission for the development of artificial general intelligence, which is beneficial to humanity, is not and will not change – even since he also said that he has not yet established what his future structure will be – but he chose to emphasize the cost of developing AI technology, which is “quite high.”

“Whatever we do, we want to strengthen this mission and this is the bar we are behaving,” he said. “The mission will not change. And in fact, the structure … I hope it will improve this mission and this is the way we think about it. “