Salesforce CEO Marc Benioff apologized on Friday for making comments in support of President Donald Trump potentially sending federal troops to San Francisco, where his company is based. 

“Having listened closely to my fellow San Franciscans and our local officials, and after the largest and safest Dreamforce in our history, I do not believe the National Guard is needed to address safety in San Francisco,” Benioff wrote in a post on X.

The Trump administration recently deployed the National Guard to Portland, Oregon and Chicago, sparking protests and lawsuits and resulting in citizens and immigrants being detained without legal representation.

In a story published late last week in The New York Times, Benioff indicated that he would welcome troops to San Francisco. The company’s annual Dreamforce conference was held in downtown San Francisco from Tuesday through Thursday of this week.

“We don’t have enough cops, so if they can be cops, I’m all for it,” Benioff told the Times.

Benioff faced blowback for his comments from local politicians and other leaders. California Governor Gavin Newsom and San Francisco politicians on Wednesday issued statements and held press conferences to deliver the message that federal troops are not welcome in the city, and that crime is coming down.

Prominent startup investor Ron Conway, who backed companies including Google, Airbnb and Stripe, resigned from the board of the Salesforce Foundation on Thursday. According to the New York Times, Conway told Benioff in an email that their “values were no longer aligned.”

Conway is a longtime Democratic donor who was a member of VCs for Kamala, and donated around $500,000 to at least two funds tied to Kamala Harris’ unsuccessful 2024 election campaign. While Benioff has donated to members of both parties, he has supported Democrats for president, including Barack Obama, Hillary Clinton and Kamala Harris.

Venture capitalist David Sacks, who is now Trump’s AI and crypto czar, said after the news about Conway that Benioff could join the Republicans. On Tuesday, Sacks, a longtime friend and associate of Elon Musk, was featured with Benioff in an onstage interview at Dreamforce.

“Dear Marc @Benioff, if the Democrats don’t want you, we would be happy for you to join our team,” Sacks wrote on X. “Cancel culture is over, and we are the inclusive party.”

Following Benioff’s initial comment to the Times, Benioff appeared to walk back his comments, writing on X that safety is “first and foremost, the responsibility of our city and state leaders.” However, by that point, Musk and other right-wing figures had seized on his original comments, amplifying them to their audiences.

Musk, who has drawn criticism for his personal drug use, characterized downtown San Francisco as a “drug zombie apocalypse.” And on Wednesday, Trump called San Francisco “a mess,” and suggested possibly sending in the National Guard.

“My earlier comment came from an abundance of caution around the event, and I sincerely apologize for the concern it caused,” Benioff wrote in his Friday post. “It’s my firm belief that our city makes the most progress when we all work together in a spirit of partnership.”

Opposition to Benioff’s initial suggestion also came from Garry Tan, CEO of startup incubator Y Combinator. He wrote on X that “We don’t need the National Guard,” but he used his post to go after liberal local officials and judges perceived as too lenient.

— CNBC’s Lora Kolodny contributed to this report.

WATCH: Benioff interview at Dreamforce

The impact of artificial intelligence is being felt across the health-care industry, from tools that help analyze medical images, aid in drug discovery and complete cancer screenings.

But increasingly, AI is being used to fight one of the biggest challenges that health care is facing: a time-challenged (and often burned out) workforce of doctors, nurses and support staff that wants to focus more on patients and care instead of other arduous administrative tasks.

That’s leading health-care technology stalwarts like Epic Systems to invest in a variety of AI features that can do everything from helping patients book appointments or understand lab results to AI-powered clinical documentation that helps doctors record and write notes in real time, as well as anticipate patient information that a doctor might need handy like blood pressure trends.

It’s also fostering a new wave of startups creating solutions for this problem. More than 60% of the venture funding that had flowed into healthcare-focused AI companies between 2019 and 2024 was focused on administrative and clinical uses, according to a report from Silicon Valley Bank.

Abridge, which ranked No. 47 on the 2025 CNBC Disruptor 50 list, was founded with the idea of “giving physicians back their time,” co-founder and chief technology officer Zachary Lipton said at the CNBC AI Summit in Nashville on Wednesday.

Self-described as “generative AI for clinical conversations,” Abridge’s platform transcribes patient-doctor conversations and uses AI to add context and data from previous patient visits and tests. That aims to save time they’d use documenting visits and instead let them focus on the patient in front of them.

“Doctors are spending two hours doing digital paperwork for every one hour of direct patient care,” Lipton said. “The status quo was that [the industry] had created a digital world where technology was taking doctors away from their patients.”

Steve Beard, CEO of health-care education company Adtalem Global Education, said at the CNBC AI Summit that “everything we know from the surveys of clinicians today is that the No. 1 driver of burnout and career dissatisfaction is the administrative burdens associated with practice.”

“All the evidence suggests that AI is a fantastic complement, and we should be encouraging the adoption of these tools because of what they can do for the experience that we all have in health care,” Beard said.

But this influx of new AI health-care tools also means clinicians need to be prepared to use them. Only 28% of physicians say they feel prepared to leverage AI’s benefits, despite 57% saying they are already using AI tools for things like ambient listening, documentation, billing or diagnostics, according to a report from health-care technology platform Inlightened.

Adtalem, which has more than 90,000 enrolled students spanning nursing, medicine and other health-related professions, recently announced a new AI credentials program with Google Cloud focused on AI applications for health-care roles. The program, which will launch next year, will also be available to clinicians at the 270 health-care systems across the U.S. with which Adtalem is partnered.

The new program will not only provide broader health-care AI fluency, but “domain-specific tools for clinicians, nurses, doctors, imaging techs and others,” Beard said.

“The technology and the pace of the development of the technology will move as rapidly as lots of other innovations have, but the critical contingency that has to be solved for is workforce readiness,” Beard said. “How do we get clinicians ready to adopt these technologies in ways that allow the investments to deliver the returns?”

Ushering in AI, regardless of industry, has raised worker anxiety around job losses, especially as CEOs have touted how the technology has upped productivity while not requiring additional headcount.

While Beard acknowledged that “every major tech innovation comes with some labor dislocations,” he said that in regard to health care, “the human element, particularly in the way it drives trust between the clinical and the patient, is something that can’t really be replicated by machines.”

“Clinicians will have a chance to do more of what they joined these professions to do in the outset, which is to be at the bedside, caring for patients,” Beard said.