At Microsoft’s annual Build conference on Tuesday, Executive Vice President Jay Parikh’s keynote was interrupted by an employee protesting the company’s contracts with the Israeli government. The protester at the Seattle Convention Center was quickly whisked away by security guards, including some undercover agents dressed like attendees.

More than 800 miles south in Mountain View, California, security guards lined the mainstage of Google I/O, where Alphabet CEO Sundar Pichai was set to speak. At the entrance to the developers conference, roughly two dozen black-clad guards rifled through bags, opening up lipstick cases and pulling out items, including women’s feminine products, and confiscating over-the-counter pain medications.

The vibe is different during this year’s tech conference season. Tensions were already elevated after the October 2023 attack by Hamas on Israel and the extended bombing campaign in the Gaza Strip that followed. But they’ve heightened in recent months as artificial intelligence technologies advance at a rapid rate and an AI arms race has entered the most sensitive parts of society.

Additionally, there’s the aftermath of the fatal shooting in December of UnitedHealthcare CEO Brian Thompson in midtown Manhattan as the executive was on way on his way to an investor event.

“We definitely have seen an uptick in the request for security, specifically in the last six to nine months,” said Richard Dossett, client relations manager for American Global Security, which works with tech companies. “There have also been a lot of protests and civil upheaval, especially in Fortune 500 companies, with the landscape at the moment, so they want extra security to make sure they’re not going to be hassled.”

Security firms and industry experts told CNBC that technology companies’ increased work with governments has contributed to an uptick in security needs. AI companies in recent months have been walking back bans on military use of their products and entering into deals with defense industry giants and the Defense Department.

Companies are responding to increased outrage in part by trying to quell internal dissent. Google last year expanded its list of prohibited discussion topics to include international issues, territorial disputes, national policy events and military conflicts.

For Microsoft, this week’s protests had recent precedent.

In April, former employees interrupted the company’s 50th anniversary celebrations, calling Microsoft AI CEO Mustafa Suleyman a “war profiteer.” Ibtihal Aboussad, then a software engineer in the company’s AI division, walked toward the stage at the event in Redmond, Washington saying, “You claim that you care for using AI for good, but Microsoft sells AI weapons to the Israeli military.”

Employees at the company had previously formed a group called No Azure for Apartheid, following the creation of similar movements at Google and Amazon directed at opposing work with the Israeli government.

Parikh, who runs the newly created CoreAi group at Microsoft, heard that specific message during his Build speech this week.

“Jay!” yelled the worker from the audience. “How dare you talk about AI when my people are suffering! Cut ties! No Azure for apartheid! 

CEO Satya Nadella was interrupted during his keynote by an employee named Joe Lopez.

“Satya! How about you show how Microsoft is killing Palestinians?” Lopez screamed. “How about you show the Israeli war crimes are powered by Azure.”

Another employee followed, “As a Microsoft worker, I refuse to be complicit in this genocide. Free Palestine!” That employee was later fired, as was Lopez.

Following the protests at Build, Microsoft employees reported that emails with the words Gaza, Palestine or genocide wouldn’t send, and expressed concern they were being blocked by the company, according to screenshots, recordings and documents viewed by CNBC.

Microsoft didn’t respond to a request for comment about the heightened security. Regarding the email issue, a Microsoft spokesperson said in a statement that some messages were being “sent to tens of thousands of employees and we have taken measures to try and reduce those emails to those that have not opted in.”

Google didn’t provide a comment about its security presence at I/O, but a spokesperson pointed to the list of prohibited items at Shoreline Amphitheater, where the conference took place.

Google had a similar situation at its developer conference last year, when dozens of pro-Palestinian protesters rallied outside with red paint on signs and clothing to signify blood. Banners and signs read “Don’t Be Evil” and “Stop fueling genocide.”

The demonstrators demanded the tech giant withdraw from its Project Nimbus contract, a $1.2 billion deal to provide AI technology to the Israeli government.

“We won’t stop ’til Nimbus is dropped,” protesters chanted.

Bombace said that as tech companies collaborate with governments, they “have to meet certain security standards.”

“We’re providing services right now in response to activity based on the conflict in Gaza,” Bombace said. Social media companies, he said, “have a whole unique footprint of issues because of the nature of their business and the things that are being posted on their platforms.”

Last year, during a keynote speech in New York from a Google executive, an employee in the company’s cloud division protested publicly, proclaiming “I refuse to build technology that powers genocide.” The hired event security forced him out of the building and the company later fired him. Google ended up terminating more than 50 employees after a series of protests against Project Nimbus last year.

Dossett said he’s also noticed an uptick in protesters trying to gain access to corporate campus buildings to record videos or take pictures to get their messages to the public. 

“When people try to invade a company’s space and film it on camera and it goes viral — that’s something other companies see and think ‘we don’t want that to happen to us,'” Dossett said. “It could affect their brand but largely, it’s been about safety of the people.”

At Build, Microsoft’s use of undercover guards plays into a growing trend, experts said.

“They’ll be in the crowd and say ‘we have a suspicious male who’s wearing a white shirt in row three,'” Bombace said. “There’s a lot that goes on that the average person doesn’t recognize and that’s good.”

It’s not just at conferences and on campuses where companies are taking extra measures for protection.

Google lifted Pichai’s security costs by 22% in 2024 to $8.27 million. At least a dozen S&P 500 companies have highlighted increased security costs, Reuters reported last month, based on an analysis of recent disclosures. Bombace said the AI arms race is a big reason for companies to boost spending in that area.

“It’s a race right now and that leads to increased security,” Bombace said. He added that to foreign adversaries, “technology becomes the No. 1 target.”

WATCH: Balancing search, AI at Google I/O