The CEO says we’re only in the “AOL” stage of AI

Current AI systems are like last year’s dial-up internet, according to the CEO of an AI startup, who said the space needed a reality check.

Sachin Duggal, co-founder and CEO of Builder.ai, told CNBC on Friday that we’re just beginning to imagine what’s possible with artificial intelligence.

“We’re still only in the AOL AI world,” Dougal told CNBC in an interview. “There is this perception that we are in the fiber optic world of AI. We are a long way from that.”

“It’s not just an LLM [large language models] And ChatGPT, although that seems to be the epicenter of how people think about it.”

The hype around AI has reached a fever pitch over the past few months on the back of excitement around ChatGPT, the popular AI-powered chatbot.

Venture capitalists are investing heavily in startups to develop AI tools in the hope that this will mark as big a shift in the digital economy as the invention of the iPhone.

ChatGPT has amassed more than 100 million users since its release in November 2022, according to investment bank UBS, making it one of the fastest growing consumer apps ever.

“AOL made the Internet easily understandable to people. BlackBerry made messages understandable,” Dougal said. “At one point it was the most popular device, and people were lining up to get the phone. It was the Apple of its era.”

He added, “What you’re seeing now is a momentum where something people didn’t understand that was esoteric became a little bit more personal.”

But, he added, the technology is surrounded by hype. “It made people freak out for no reason.”

ChatGPT has impressed many with its ability to produce human-like responses to user prompts with the support of large language models trained on huge amounts of data.

However, it has also proven ineffective for some tasks, such as solving math problems. The chatbot also has a limited understanding of context – especially sarcasm and humor.

Cognitive graphs — data models that connect relationships between different concepts, entities, and events — show a greater degree of fidelity and contextual understanding than large language models like OpenAI’s GPT-4, Dougal said.

“LLM is simply telling you what it thinks the next word is with a high degree of probability, whereas the knowledge graph is actually able to make pattern relationships that it knows, how things go. So it’s not just about predicting what’s next,” he said.

He watches: AI allows a more creative part of human nature to begin: Builder.ai CEO