OpenAI chairman Bret Taylor lays out the bull case for AI agents

Spread the love

We have not yet received the direct definition of its Exactly the key to the AI ​​agent Fireside chat in Barcelona on Tuesday during Brett Taylor’s Mobile World Congress. The Sierra The founder and the OpenAI board chair side-steps CNN moderator asked the question of how “agent AI” “Jenny is different from the chattabot” to hate everyone, but AI agents are delighted to see “sympathetic” reacts to “sympathetic”.

His new startup is creating a Customer Service AI agent, you hope that the gospel about the potential of Taylor technology will be promoted. And he is not disappointed: “I am more interested in big language models and I am more ecstatic than the current technology I can think of, probably since I discovered the Internet in adolescence,” he told the conference representatives.

The change of steps with the previous repetitions of the AI ​​chattobs is compared to the AI ​​chattobs in the generator AI-Jalani Customer Service AI.

“I think we are now in this era where these AI solutions are actually better than the options,” he added: “We work with companies like SiriusxMM in the United States, or EDT Home Security, where your alarm stops working on it will help you fix it, and no need to wait for a field service team in your home.

“And what is notable about these agents is that people actually like them”

Supercharging customer experience

These more capable AI services bots are helping companies to shrink customer service costs, which Taylor suggested that the bar would help increase overall. “I think it’s just going to improve consumers’ experiences of many brands,” he said.

The bottles that are very capable can lead to new challenges, though he acknowledges the examples where customer support AI agent does not exist in a customer mourning “Hallucinated” refund policies.

He said that the “Guard Rail” developed brands suitable for their AI agents thus an important part of the equipment safely implemented, he said. However, he was a bullsh that the customer service agents will shrink this challenge as it is growing for the use-case and policies of each brand.

“In general, my philosophy is, do not wait for the technology to be perfect. In fact, it can never be perfect – but narrow the domain you are working with so that you can take these complex problems and make them solved, “he said.

“Instead of trying to solve all the AI ​​problems in the world, you compress it in a domain, ‘Hey, we’re going to set up some practical guards around this AI so we can solve the problems right now.’ And I think this is an opportunity for each organization at this conference, ”he said. In addition to the AI ​​Agent Company focusing on its own customer service, he has checked the name AI code assistant Cursor And openAI-supported legal technology Harvey AI is an example of specialization that is successfully applying AI agents in a defined domain.

Taylor’s adoption of how to be a seminal AI agent for brands in the future was amazingly maximal. “I think most companies, AI agents will be as important as their website or their mobile app in the per cent of the interactions they have with their customers,” he said. “It won’t surprise me for most brands here if in reality, if you are five or 10 years ahead, their AI agent is their main digital experience, which I think is the same difficult to imagine right now. But I really thought that the world was going there “” “

How people communicate with AI agents can probably be transferred, he further suggested that the user interfaces would become more faded in the background because technicians seek more effortlessly to tap technology utilities.

“I think it’s – I’m hopeful – all the time that everyone looking at their screen will start melting as a social habit. And with the advent of the conversation AI, when the software really understands how we talk, computers will be sorted and the devices will melt some kind of melting and I think it will be very exciting, “he said as a parent. As a parent, he said he would not have to look at his own children” to engage in their children’s technology. “

Racilling

What about customer service AI agents’ hindrance can be done?

Taylor said that it was a legitimate concern but again expressed optimism that the shift would eventually be good for humanity – hoping that some of the work role would go away, the new ones would be exposed in their place. However, he added that “Technology makers have a responsibility to do that conversation and just not supply the technology.”

He said that the big risk with the AI-fascinating job shift will not be able to keep pace with the required riskling change rate, he said. “It is a disruptive force when society can resurrect. So basically, I think it requires official, personal partnership ””

The moderator also asked the OpenAI Board chair to switch to a non -profit initiative about the AI ​​giant plan, which has attracted some critical attention.

Taylor said that the mission of development of the artificial general intelligence of the openness that does not benefit humanity does not benefit – even he also said that it was not yet fixed what the structure of his future – but he chose to highlight the cost of developing AI technology, which he said “very high”.

“Whatever we do, we want to make that mission more wide and this is the time that we hold on to ourselves,” he said. “The mission will not change. And in fact, the structure … I hope, to enhance that mission and we are thinking about it “”

Leave a Reply

Your email address will not be published. Required fields are marked *