Marc Benioff, CEO of Salesforce, sat down with Bloomberg in the latest in a spate of video interviews and podcast appearances.
These appearances have been transparent in their aim: to promote Agentforce, even if that means trashing industry rivals (*cough Microsoft).
This time around, during the episode of Bloomberg Originals, the host quizzed him on a running joke about the general public not knowing what Salesforce actually does.
Again, he masterfully brought attention back onto Agentforce by recalling a text exchange with Elon Musk, the Owner of Tesla and X.
Musk questioned why the company was called Salesforce when it has such a wide range of products that expand beyond the sales sector.
In response, Benioff quipped: “What should we call it? Agentforce?”
While it seems like the CEO has opted against the name change, the comment highlights how hard Salesforce is pivoting toward Agentforce, and how the platform is dominating the company’s go-forward strategy.
The mentioning of the platform naturally led to further discussions about Salesforce’s overall AI approach, with Benioff stating that “AI is doing 30 to 50 percent of the work at Salesforce now. And I think that will continue.
I think that we all have to get our heads around this idea that AI can do things that before, you know, we were doing. And we can move on to do higher-value work.
When pushed on whether he had any ethical qualms about promoting AI solutions that could be used to replace human labor, the CEO implied that the power of the tech made it inevitable, even joking that he “hopes” an AI agent will be able to replace him.
“We’re probably looking at $3 to $12 trillion of digital labor getting deployed. And that digital labor is gonna be everything from AI agents to robots,” he explained.
Benioff’s AI Concerns: From Monopolies to Security
While Benioff did spend much of the discussion waxing lyrical about AI agents, he also outlined some of the challenges and issues surrounding the tech.
Most notably, he claimed that the power of AI strengthens the argument for breaking up big tech.
In particular, he was referencing the recent trials of Google and Meta, where the companies have been accused of claiming monopolies, explaining how, from an AI perspective, these companies that make their own models are gaining control over entire industries. He said:
For companies that control or have monopolistic aspects of their business, it [the break up of big tech] has to be looked at.
“I think that’s a key part of the government’s role when it comes to competition. And that’s definitely one of the guardrails of capitalism.”
Another common AI concern discussed by Benioff was accuracy and security.
The CEO admitted that the tech is not yet 100 percent accurate, stating that Salesforce is currently operating at around 93 percent.
“You can never have extreme confidence in security. You have to be completely paranoid,” he said.
When asked whether 93 percent is good enough, Benioff claimed it was “pretty good” and a lot higher than other vendors in the space.
According to the CEO, these can’t compete with Salesforce’s data and metadata capabilities, but did not expound on whether the standard was currently up to par or how the company intends to close the gap to 100 percent.
The Benioff-Microsoft Beef is Alive and Well
One of the reasons that Benioff’s name has never been too far from the headlines in the last few months is due to his series of attacks on Microsoft and its Copilot solution.
In fairness to the Salesforce man, this time around it was the interviewee who brought up Microsoft, but he appeared to relish another opportunity to tear into his biggest rival.
On the weekend of the Glastonbury Festival, it was quite apt that Benioff’s remarks played out like a greatest hits set.
He once again described Microsoft Copilot as “Clippy” and “repackaged ChatGPT”, as well as reiterating his belief that users were “disappointed” with the tool.
However, he also provided insight into why he has felt the need to continuously take shots at Microsoft over the past few months.
“I think we had to make it clear, the difference between what we were saying and what Microsoft was saying,” he explained.
What Microsoft was saying was, ‘Hey, this is a Copilot revolution. We’re all gonna have this Copilot. It’s about the Copilot.’ And then customers were not connecting with that.
“We said, no, look at this; this is the Agentforce vision; this is what AI was meant to be.
“These are agents who are making you more productive, reducing your costs, and making things easier,” he concluded.
Elsewhere, Benioff recently celebrated the launch of Agentforce 3, with Salesforce establishing a Command Center for managing enterprise-wide AI agent implementations.
The move underlines the broader industry pivot from AI agent products to AI agent ecosystems.
