Newsletter

• Proton Users are claiming to be unsubscribing and demanding that CEO Andy Yan ( Andy Yen ) resign.
• It is reported that while promoting the faster adoption of AI, Yan refers to increasing security testing requests as 'thinker of losers'.
• The company subsequently canceled the originally scheduled re-evaluation of employees using AI and issued a statement to defend its engineering reforms.

After relevant disclosures showed that CEO Andy Yan pressured engineers to adopt AI and dismissed requests for increased security testing as "loser mentality," Proton users are reporting that they have unsubscribed.

This Swiss privacy company has since canceled the original assessment for its employees AI.

This rebound has hit one of the core selling points of Proton – namely, convincing users that they can entrust information they wish to keep confidential to it. Its encrypted email service is supposed to protect the content of emails, but now customers who question its development process are asking whether a faster release pace might come at the cost of their interests.

According to 404 Media, last August, an employee Proton requested a few days from the company to conduct penetration testing on the new code, which means having experts try to break into the code before its release. Yan replied, "That's nonsense; it's the mindset of a loser. Of course, speed and security can be achieved simultaneously."

The report states that he also urged employees to stop writing code by hand and to use more of AI, and mentioned that he would re-interview the employees to understand whether and how they use AI to complete their work. The media said that its investigation was based on interviews with 9 current and former employees, internal information, and meeting recordings.

This approach has not been well-received by users. Customers have responded on social media platforms such as Reddit with requests to cancel subscriptions and calls for a change in leadership. On the feedback forum of Proton, a proposal demanding their immediate resignation has received 2,020 votes during review, and comments from users also indicate that they have already cancelled their subscriptions or will not renew them in the future.

In addition to the blind use of AI, some current and former Proton employees have also documented a toxic work culture on Reddit and social media, with particular criticism directed at CEO.

However, these posts record angry emotions, rather than the verified number of customers who have already been lost. Canceling a paid subscription does not necessarily mean deleting an account.

The ambition of Proton towards AI predates this controversy. Its Lumo chatbot comes with protective measures to prevent conversations from being used for training AI. However, even during the peak of the chatbot craze, the internal use of AI was not one of the selling points for Proton's external promotion.

In response to concerns from the public about the irresponsible use of AI, on November 1st, Bart Butler (Bart Butler), the technical director of Proton, issued an engineering manifesto advocating that developers should take control of large language models—those systems capable of generating text and code—rather than treating them merely as occasional assistants.

Butler compares writing code manually to memorizing phone numbers and envisions eventually eliminating most manual code reviews. However, he also insists that professional skills and testing are still essential, as engineers must still be responsible for tasks that models cannot handle reliably.

The declaration states: "Your job has fundamentally changed. It's no longer about writing code, and LLM is not there to help you write better code more quickly either." "Your job is to help LLM write better code more quickly." Proton further emphasizes that avoiding AI poses a 'survival threat' to the success of its mission itself.

“‘Programming’ is dead. Engineering is not dead,” Buttler said to his team, “Your previous jobs no longer exist.”

The document also calls on companies to invest in automation across the board, stating that if approval and other development processes remain slow, then the ability to generate code more quickly is of little significance. Butler describes this reform as the next phase in adopting AI.

AI The tools themselves can also introduce security issues. For example, Microsoft researchers have indicated that a vulnerability, now fixed (Claude), could potentially expose secret access credentials through malicious instructions hidden in materials reviewed by coding assistants.

But AI can also help in identifying weaknesses: Google's PageBreak system searches for exploitable vulnerabilities within its applications and checks whether the attacks are actually effective.

Proton admitted to Media that sometimes the communication from management is too rigid, which also causes anxiety among employees. Nevertheless, the company's publicly issued statement still believes that if AI is not adopted quickly, it will threaten their competitiveness.

As of October 30th, the re-interview arrangements were cancelled. Originally, these assessments would have required employees to explain how they used AI, and employees were concerned that their answers might determine whether they could keep their jobs or not.