During Bret Taylor’s fireside chat at the Mobile World Congress in Barcelona on Tuesday, a clear definition of an AI agent remained elusive. The Sierra founder and OpenAI board chairman sidestepped CNN moderator Anna Stewart’s inquiry regarding the distinction between “agentic AI” and a GenAI chatbot. Instead, he suggested that people tend to dislike the former while enjoying the “empathetic” responses offered by AI agents.
As his new venture is focused on developing a customer service AI agent, one might expect Taylor to be enthusiastic about the possibilities of the technology, and he delivered. “I am more thrilled about large language models and this latest wave of innovation than I have been for any technology since I discovered the internet as a teenager,” he shared with the audience.
The significant advancement in generative AI-driven customer service agents compared to previous AI chatbots lies in their vastly improved capabilities, including features that allow them to be “multilingual and instantaneous.”
“We are entering a new era where these AI solutions are often superior to previous alternatives,” he stated. He highlighted their partnerships with companies like SiriusXM and ADT home security, explaining how AI can assist consumers without the delays of needing a technician when issues arise.
“Remarkably, people actually appreciate these agents,” he noted.
Enhancing the Customer Experience
These advanced AI service bots are enabling businesses to reduce customer service costs, a trend Taylor believes will elevate the overall consumer experience across numerous brands.
However, he acknowledged that highly capable bots can introduce new challenges, citing instances where customer support AI agents have “hallucinated” responses about refund policies during sensitive situations, such as customer bereavement.
He stressed the importance of developing appropriate “guardrails” for AI agents to ensure their safe implementation. Despite this, he remained optimistic that as customer service agents are increasingly customized to fit individual brands’ needs and policies, these challenges will diminish.
“In general, my philosophy is not to wait for the technology to become flawless. In fact, perfection may never be achieved — but limit your focus to a specific domain, allowing you to tackle complex problems and find solutions,” he advised. “Rather than attempting to solve all of AI’s challenges worldwide, hone in on a specific area and establish practical guardrails around your AI to address immediate issues. This presents an opportunity for every company here,” he added, citing AI code assistant Cursor and OpenAI-supported legal technology provider Harvey as successful examples of specialized AI applications.
Taylor’s vision of the importance AI agents will hold for brands in the future is strikingly ambitious. “I believe for most companies, AI agents will become as pivotal as their website or mobile application in terms of customer interaction,” he suggested. “Fast forward five or ten years, and it’s conceivable that for many brands, their AI agent will represent their primary digital experience. That might seem hard to fathom now, but I genuinely believe that’s the direction we’re heading.”
He also predicted a transformation in how users engage with AI agents, envisioning that user interfaces will become less prominent as developers strive to make interactions with the technology even more seamless.
“I’m hopeful that the habit of everyone constantly staring at their screens will gradually diminish. With the rise of conversational AI and software that comprehends human speech, I anticipate that computers and devices will become less noticeable, which I find very exciting,” he remarked. As a parent, he hopes that his children “won’t have to be glued to screens their entire lives to engage with technology.”
The Responsibility for Reskilling
What does the rise of customer service AI agents mean for employment?
Taylor acknowledged the concern but remained optimistic that the overall shift would benefit humanity. While some job positions may disappear, new opportunities should arise in their place. He emphasized that “technology creators have a responsibility to engage in this dialogue rather than simply releasing new technologies.”
He warned about the risks of an AI-driven workforce transition outpacing necessary reskilling efforts, stating, “When disruption occurs faster than society can adapt, it becomes a disruptive force. This fundamentally requires cooperation between the public and private sectors.”
When asked about OpenAI’s transition from a nonprofit to a for-profit entity, which has stirred some controversy, Taylor reaffirmed that the company’s mission to develop artificial general intelligence for the benefit of humanity remains unchanged, even though its future structure has yet to be determined. He emphasized, however, the high costs associated with AI technology developments.
“Whatever course we take, we are committed to amplifying our mission, and that is the benchmark we hold ourselves to,” he stated. “Our mission will endure. Moreover, I hope the structural changes will further enhance that mission; that is our current thinking.”
Compiled by Techarena.au.
Fanpage: TechArena.au
Watch more about AI – Artificial Intelligence
