Proton is confronting a wave of subscription cancellations and calls for Chief Executive Andy Yen to resign, after revelations that he pressured engineers to embrace artificial intelligence and dismissed a request for additional security testing as "loser thinking."

The Swiss privacy company has since canceled planned employee assessments of AI use, but the backlash strikes at the heart of its core selling point: being trusted with information people want kept private. Its encrypted email service scrambles messages to protect their contents, and customers now question whether faster releases could come at their expense.

The 'Loser Thinking' Exchange

According to an investigation by 404 Media, a Proton employee asked the company in August for a few days to pentest new code — meaning specialists would try to break into it before release. Yen replied: "That's bullshit and loser thinking. Of course it's possible to be fast and secure."

Pre-release testing of that kind is a standard safeguard in software engineering, and it carries particular weight for a company whose paying customers are buying a security promise rather than just an inbox.

He also urged staff to stop writing code by hand and use more AI, and said he would re-interview employees to determine if and how they were using AI to do their jobs, the outlet reported. The investigation drew on nine current and former employees, internal messages, and meeting recordings.

Users Call for New Leadership

The report triggered immediate backlash, with customers posting cancellation notices and demands for new leadership on social platforms including Reddit. A proposal on Proton's feedback forum calling for Yen to step down had gathered 2,020 votes when reviewed, alongside comments from users saying they had canceled subscriptions or would not renew.

Beyond concerns over blind AI adoption, some current and former Proton employees have taken to Reddit and other social media to document a toxic work culture, particularly from the CEO. Those posts record anger rather than a verified count of departing customers, and canceling a paid plan does not necessarily mean deleting an account.

Manifesto Doubles Down on AI

Proton's AI ambitions predate the dispute. Its Lumo chatbot comes with protections against using conversations to train AI, but internal use of AI was never among the company's selling points, even at the peak of chatbot hype.

To address concerns over irresponsible use of AI, Proton's technology chief Bart Butler published an engineering manifesto on November 1, arguing that developers should direct large language models — AI systems that generate text and code — rather than treat them as occasional assistants.

Butler compared manual coding to memorizing phone numbers and envisioned eventually ending most human code review. He also insisted that expertise and testing remain essential, with engineers accountable for work that models cannot reliably handle.

"Your job has fundamentally changed. It is no longer to write code, and LLMs are not here to help you write better code faster," the manifesto reads. "Your job is to help the LLM write better code faster."

Doubling down on its stance, Proton sees avoiding AI as an "existential threat" to the success of its mission. "'Programming' is dead. Engineering is not," Butler told his own team. "Your old job is gone."

The document also calls for company-wide investment in automation, arguing that generating code faster achieves little if approvals and other development steps remain slow. Butler described the overhaul as the next phase of adoption. For subscribers, the practical question is how that overhaul will square with the testing practices at the center of the current dispute.

AI Tools Cut Both Ways on Security

AI tools can introduce security problems of their own. Microsoft researchers said, for example, that a now-patched Claude flaw could expose secret access credentials through malicious instructions hidden in material the coding assistant reviewed. They can also help find weaknesses: Google's PageBreak system searches its applications for exploitable bugs and checks whether attacks actually work.

Those episodes reflect a tension now running through the wider software industry, as companies push AI deeper into development while security researchers document how the same tools can be turned against the systems they help build.

Proton acknowledged to 404 Media that management had sometimes communicated too bluntly and that employees were anxious. Its public manifesto nevertheless argues that failing to adopt AI quickly threatens the company's ability to compete.

Yen canceled the re-interviews on October 30. The assessments would have required staff to explain their AI use, with employees fearing their answers could determine whether they kept their jobs. The manifesto followed two days later, meaning Proton's public case for rapid AI adoption remained in place even as the re-interviews were called off.