Proton's CEO cancelled mandatory AI "re-interviews" a day after 404 Media reported staff feared firing
Employees expected to be questioned on how much they use AI, and believed those who didn't use enough could lose their jobs.
Six current employees and audio of a two-hour all-hands told 404 Media the same thing: the sessions were a test, and failing it could cost a job. Andy Yen called them off, admitting a "communication failure on my part" while insisting the AI push continues.
He also promised a reorganisation next month with "no loss of jobs from AI". Staff told the reporters the criteria may still land in performance reviews.
What 404 Media found, and what Yen said at the all-hands
404 Media reports that Proton CEO Andy Yen cancelled a near company-wide set of "re-interviews" on Friday. The sessions would have required employees to explain how they use AI. The account rests on six current employees and audio of a roughly two-hour all-hands.
The reversal followed an investigation published the day before, drawing on nine current and former staff, Slack messages, meeting audio and an AI manifesto from CTO Bart Butler that told engineers, "Your old job is gone." In one Slack exchange Yen rejected a suggestion to pentest new code before release: "That's bullshit and loser thinking. Of course it's possible to be fast and secure." Employees described holding back security concerns because Yen mocked people in front of colleagues.
At the all-hands Yen acknowledged a "communication failure on my part" but said the re-interviews had already done their job by getting staff to learn the tools. He announced a reorganisation for next month with "no loss of jobs from AI". Butler admitted the company had made a mistake but said AI use did not mean dropping security. Some employees welcomed the change; others called it damage control, noting that the criteria would reportedly feed into performance reviews. Yen's Reddit reply collected more than 1,200 downvotes, with users saying they were cancelling paid plans.
The thread: "That 'loser thinking' line should never be accepted"
softwaredoug took the quote personally:
"That "loser thinking" line should never be accepted in any company discourse, even from leadership. If this person wasn't an exec, they'd be shown the door for that language."
A first-round investor at Uncommon Capital disagreed with the framing of the complaint:
"It's a team not a family. The idea that demanding performance from your employees is "losing your humanity" is insane."
A commenter read the episode as a labour story:
"Techies are getting a glimpse at how other employees have always been treated all along, and they don't like it."
Customers are shopping for an AI-free mailbox
The practical fallout showed up in the replies. One reader wrote "I used to think Proton was good" and said they were cancelling. Another was hunting for a replacement:
"I'm looking at alternative email provider options, and I'm struggling to find options that have banned GenAI, and have the features and reliability I need. I will gladly pay a premium for a provider that's not going to let AI creep in gradually, and take a bigger role over data and security"
A Fastmail user of several years offered the standard answer: "I haven't seen any AI fluff yet…". For a company whose product is trust, the price of an internal management fight turned out to be measured in canceled subscriptions, not morale surveys.
Get breakdowns like this every day
The top Hacker News stories, unpacked — by email, Telegram, or RSS.
Northflank: assume every environment variable is stolen, and customers say AI keys are already being used
The status page went from "environment variables were decrypted" to "assume all of them have been compromised" within a day.