For business executives, A.I. is a force that can bring customers closer, or take them away

“I unfortunately became the poster child for how AI can fuck you,” said Rosensweig at a lunchtime panel at the Fortune Brainstorm Tech conference in Deer Valley, Utah this week. In Rosensweig’s telling, he “made the mistake of telling the truth” about losing 100,000 paying Chegg subscribers with the launch of free service ChatGPT 4 in April. “I don’t think anybody’s A.I. ready; I don’t think anybody knows what ‘A.I. ready’ means.”

While Rosensweig highlighted a downside risk of A.I., other executives on the panel were cautiously optimistic about the technology and its potential to help them forge stronger relationships with customers.

Amit Patel, CEO and president of $8.5 billion (market cap) discount shopping platform Rakuten believes a most-promising use for A.I. is in checkout. “Based on the data that we have, the checkout piece still remains the biggest friction point throughout the whole experience,” he said on the panel, alongside Rosensweig, Cotopaxi chief brand officer Brad Hiranaga, Salesforce’s president of global strategic customers and partners Jim Steele, and Fortune moderator Michal Lev-Ram. Patel says that $20 billion is abandoned in Rakuten shopping carts annually, and believes that upwards of 10% of this can be converted to sales via A.I. “Whether it’s Apple Pay, buy-now-pay-later companies, Google Pay, you name it, all of these things have their own intricacies that only work on certain platforms. Our goal is to figure out how do we make it fast? How do we make it easy? How do we make it joyful?”

For leaders at other companies—organizations both larger and smaller than Rakuten—basic questions abound about where, and how, to utilize A.I. in the customer-oriented parts of their businesses. “I think it’s a net positive, but in terms of understanding what it’s going to be for our business, it’s still very questionable,” said Hiranga, who oversees 311-employee (per LinkedIn) cult outdoor apparel brand Cotopaxi. 

Steele, who watched Salesforce grow to more than 72,000 employees during his more than 20 years at the comapny, echoed the sentiment. “It’s this race to figure [AI] out,” he said.

Steel said that his team has focused on determining an A.I. rollout plan, which has three parts: The first is separating sensitive data from data to be used in the large language model. The second, setting clear governance. And the final piece is ensuring employees are A.I. ready. “It’s a challenge,” he admits. “We manage more customer data than any company on the planet.”