# Will the recent Proton news (AI hysteria, toxic workplace, perception over security) affect recommendations? **URL:** https://discuss.privacyguides.net/t/will-the-recent-proton-news-ai-hysteria-toxic-workplace-perception-over-security-affect-recommendations/41842 **Category:** Questions **Created:** 2026-10-29T20:14:09Z **Posts:** 36 ## Post 1 by @ZenByte — 2026-10-29T20:14:09Z I made a separate post for this since the other threads are more focused on protest than end-user actionable recommendations. > **Other threads:** > > [Web browser, an alternative to Signal, Proton Phone: “It’s all inevitable” – Interview with Andy Yen](https://discuss.privacyguides.net/t/web-browser-an-alternative-to-signal-proton-phone-it-s-all-inevitable-interview-with-andy-yen/41358) > > [https://discuss.privacyguides.net/t/manifestos-fear-and-loser-thinking-inside-proton-s-ai-reckoning](https://discuss.privacyguides.net/t/manifestos-fear-and-loser-thinking-inside-proton-s-ai-reckoning) Concerning for me is the news that 90℅ of Proton is already written with AI. I’m not inherently anti-AI if it’s used responsibly. But… > Andy Yen, the CEO of Proton, had a message for an employee in the company’s security Slack channel who urged the company to at least take a few days to pentest newly written code before pushing it live. “That’s bullshit and loser thinking. Of course it’s possible to be fast and secure,” Yen wrote in the channel at the end of August. “Stop thinking always of how to be slower, and focus more on how to be faster.” > > In a follow-up message, Yen reiterated his point: “only losers think like that.” His ultimate message for the team was to “stop lovingly handcrafting all our code and use AI more.” How bad is this? Is this existentially concerning from a recommendation point of view? --- ## Post 2 by @anonymous644 — 2026-10-29T20:51:35Z > [@ZenByte](#): > > Is this existentially concerning from a recommendation point of view? No :man_shrugging: I don’t want to act like audits are a panacea but I think its worth tempering our concerns over these kind of things when Proton is regularly coming out clean in external audits and fixing the issues that are reported. I think Andy Yen says a lot of stupid things that hurts the image of his product but I don’t think the product is actually as concerning as the things he says. If that makes sense… --- ## Post 3 by @blibly — 2026-10-29T21:10:14Z If they’re only just now going all in on AI, it might take time for problems to energe. --- ## Post 4 by @Boris — 2026-10-29T21:44:24Z > [@anonymous644](#): > > I think Andy Yen says a lot of stupid things that hurts the image of his product but I don’t think the product is actually as concerning as the things he says. If that makes sense… “Perception is more important than truth, the truth doesn’t matter if you don’t win on perception.” - Andy Yen --- ## Post 5 by @Sorge — 2026-10-29T22:51:01Z > [@ZenByte](#): > > Concerning for me is the news that 90℅ of Proton is already written with AI They don’t state that. 90% of the code produced in the last month was generated, but they don’t rewrite the whole thing every month. IDK if it matters, as it shows the direction anyway. I don’t think any other usable project has a different direction though. --- ## Post 6 by @velsolus — 2026-10-30T04:07:44Z This is very concerning. I’m someone who supports AI coding for efficiency, but the careless attitude about this and how sudden the shift feels brings into the question protons whole reputation of being high quality and privacy and security respecting --- ## Post 7 by @Gopher — 2026-10-30T07:09:02Z If the guy overseeing the product is sending around memos ordering his workers to play it fast and loose with security, why should we trust their product? --- ## Post 8 by @object2598 — 2026-10-30T10:19:03Z I think we should bump down in the order of recommendations, but keep them and not put a disclaimer. It’s still a net improvement compared to Big Tech, but we should probably give incentives to switch to independent, less problematic alternatives --- ## Post 9 by @Mxyzptlk — 2026-10-30T13:35:26Z The head of the company has basically said that security is secondary (or tertiary) to profit and appearance. The company puts spyware on their own workers’ systems. I’m not sure how this can align with the stated goals of this site…? --- ## Post 10 by @trilobyte — 2026-10-30T14:51:46Z When it comes to email and VPN, they’re still an established provider who are among the most secure, most private options there is. Even if this changes in the future, we are going to need more than internal memos as evidence. I will continue using them for VPN, but feel comfortable with my decision to move away from them with email. I’m definitely watching how this develops. --- ## Post 11 by @Menkork — 2026-10-30T14:57:19Z Well, this is certainly interesting timing. I woke up this morning to discover that Proton had quietly added another 2 TB of cloud storage to my Visionary account, bringing the total to 8 TB. Apparently, Visionary subscribers are also getting access to new ultra high VPN servers, and Lumo is undergoing some pricing and plan changes, with Visionary members receiving the highest non-business tier. Now, I’m wondering whether all of this was already scheduled to roll out today or whether Proton decided to push some of these benefits out ahead of schedule, given everything that’s happened over the past couple of days. I’m not suggesting there’s necessarily a connection, but the timing certainly caught my attention. --- ## Post 12 by @Learning — 2026-10-30T15:01:44Z > [@Boris](#): > > “Perception is more important than truth, the truth doesn’t matter if you don’t win on perception.” - Andy Yen Could not agree more! Learned this lesson the hard way during my career. I have way too many examples of this being the absolute truth. It didn’t matter if the facts proved otherwise, once the perception was out there, there was really no way to recover from the damage. Hard, hard lesson on this one. --- ## Post 13 by @ValiantSparrow53 — 2026-10-30T17:29:48Z Most Proton apps are open source, and they regularly pass audits. I think for now its best to wait and see if they continue to pass public scrutiny while preparing for the worst and looking into alternatives (though I’d hate to splinter everything into a dozen services). --- ## Post 14 by @velsolus — 2026-10-30T20:14:35Z The issue is that nothing can be 100% guaranteed safe, and unreviewed or untested AI code is bound to introduce some security holes --- ## Post 15 by @Sorge — 2026-10-30T20:22:01Z “Unreviewed or untested AI code” is not something any adult practices. It isn’t a serious or good-faith assumption that Proton pushes unreviewed or untested AI code, and nothing in the 404 Media materials points to that idea either. --- ## Post 16 by @Polymer7229 — 2026-10-30T22:44:38Z > [@Sorge](#): > > unreviewed or untested AI code > “I think by the end of this year most developers stop programming and start writing specs for tests that AI will code and run, **to validate** AI written code that **humans will no longer be able to review**. This will be the last year that we hire **engineers** , and starting next year we hire what we used to call **product managers** , to build things with AI.” I think it’s clear that he envisions having no human-reviewed code nor human-reviewed technical/architectural design. When do product managers review those kinds of things? AI testing _may_ be next on the menu. Why have it when the users are already doing it? I would be rethinking about my “career” with Proton if I were the engineers involved in such activities. Although it’s clear that you have to do something different across the industries, having the fear of … put into you may not be the most productive for some. It’s also clear that the CEO trusts the AIs more than you. > That’s bullshit and **loser thinking**. > > Our reporting reveals a **culture of fear** inside Proton, where many employees are hesitant to even mildly critique Yen’s views or discuss **engineering problems**. Workers describe an environment where employees have genuine **concerns about security** or being asked to develop security products with AI, but feel unable to voice them, lest Yen berates them in front of everyone or threatens their employment at the company. > > Our reporting also paints a picture of Proton that is likely very different to the **public’s perception** of a company with a **goal of carefully protecting privacy** : one that in reality is barreling ahead with AI as quickly as possible so **it can achieve the explicit goal of more annual revenue.** When security and privacy are put in a lower-priority position (they are probably never first priorities in the first place), are they going to get better or worse? When you can’t even discuss them because of the bullying/groupthink (losers!), you have to trust the AI they’re using to provide these things. I agree that we still don’t know what will happen to their products and code/design (that no human will be able to review), but we are being coaxed into the culture “trust the AI, bro.” --- ## Post 17 by @Sorge — 2026-10-30T23:14:24Z I’m not sure if this comment was actually addressed to me. If it was, my take was that Proton unlikely pushed unreviewed or untested AI code. I didn’t state that the code is reviewed by humans, and I’m not sure human review is the best option we have now, either generally or for AI-generated code specifically. If 90% of the code in October was generated by AI, human review won’t work too well. Even here, I think most people would agree that humans are pretty bad at reading AI code (but the same could not be said about AI reading human code). > This will be the last year that we hire **engineers** , and starting next year we hire what we used to call **product managers** , to build things with AI.” Right, but read in the context of “your job is dead,” it doesn’t mean hiring people without IT and security understanding. It means hiring people who get more out of AI and don’t necessarily have a purely engineering skillset. Doesn’t sound crazy to me. > Workers describe an environment where employees have genuine **concerns about security** It’s also clear from the other side, the CEO and CTO, that they’re tired of devs using security as an excuse for being slow. I think we should consider both views and not cite only one side. If you build a very secure app but it’s too slow to ship and sell to pay the bills, nobody will use it, because the company will die and the devs will be fired anyway. I’m not saying that’s Proton’s case, just using it as an example of when not prioritizing security above everything is a valid choice for both the company and users. For the company, because it wants to keep existing, and for users, because they risk losing the apps they rely on. From 404 Media and all of Yen’s quotes about “Proton is underperforming,” it looks like his logic is exactly like this. And again, all the security concerns come from the devs’ quotes. 404 Media doesn’t provide any material showing the CEO or CTO actually asking to deprioritize security. Which is kind of weird: a lot of data leaked, a lot was published, and nothing beyond the devs’ words confirms these priorities. From my experience, I have a lot of messages from employers saying “we don’t do this because it’s too expensive/long/not important.” So I’d expect to see some proof and documents in such cases, not anonymous words. Otherwise it sounds like a personal dev opinion (see the part about devs seeking excuses). --- ## Post 18 by @Sorge — 2026-10-30T23:25:15Z > [@Polymer7229](#): > > but we are being coaxed into the culture “trust the AI, bro.” I hope the culture is “let’s learn and build tools to make AI more solid.” A lot is happening in education, companies, and the open-source community around this. I’m not aware of a “trust the AI, bro” culture. --- ## Post 19 by @Boris — 2026-10-31T00:43:47Z At the moment I’m thinking about requesting my discretionary refund (6 months into my second 2-year prepaid Proton Unlimited plan) and migrating elsewhere. It doesn’t matter to me what the outcome of this drama is, or who’s in the right, what matters is that it happened in the first place. Regardless if the CEO/CTO suffer from AI psychosis or the Bluesky gang is staging a mutiny, they are not operating as a unit and it doesn’t feel like the ‘general is in control’, so to speak. I don’t want to “consider both views” or wait for their official responses or actions, I want stability in my productionised email use, what’s what I’m paying them for. If I wanted a headache, I would have self hosted email :joy: Having to keep up with Proton’s drama feels the same - unnecessary work (maybe it’s fun from a privsec aficionado perspective if you’re not actually relying on the service for some critical admin). Anyway, how can I be sure that a rogue Proton employee group won’t do something malicious with Proton Pass as some form of rebellion in the future? Their control environment clearly wasn’t strong enough to prevent the 404 Media article from happening, which is my primary concern. --- ## Post 20 by @trilobyte — 2026-10-31T14:59:57Z > Regardless if the CEO/CTO suffer from AI psychosis or the Bluesky gang is staging a mutiny, they are not operating as a unit and it doesn’t feel like the ‘general is in control’, so to speak. You should probably consider that, with a smaller group, you will never hear about it if something like this is happening. Proton is big and under a lot of scrutiny, so you find out about most of their drama. On its own, it’s not evidence Proton has more drama going on. On the other hand, if I find out my beloved and mom and pop grocer is run by an owner who makes of habit of demeaning and insulting their employees… I would just not go there anymore. No reason for it to be different with a company’s CEO and main share holder. Other companies I’m sure also have bad managers, but it’s a bad sign when it’s the CEO himself getting his hands dirty and harassing employees. --- ## Post 21 by @anonymous644 — 2026-10-31T18:22:03Z A very similar quote is also attributed to Ivanka[Trump](https://m.economictimes.com/us/news/quote-of-the-day-by-ivanka-trump-perception-is-more-important-than-reality-if-someone-perceives-something-what-her-controversial-advice-says-about-success-image-and-power/articleshow/133800870.cms) --- ## Post 22 by @Boris — 2026-10-31T19:32:18Z That’s why I’m thinking about it and haven’t made a decision yet, but even so: why did they scale inorganically so fast to the point they cannot maintain a good control environment? What happened - happened, what does the good ending look like from here? The CEO stepping down or being removed due to external pressure? All the ‘rebels’ getting fired and the CEO cleaning up? I honestly just wanted an email provider to do my emails, only reason I upgraded to Unlimited was to use more custom domains (3 in Proton and am forced to use 6 in SimpleLogin). I even contemplated upgrading to Visionary to support the company and use more custom domains and aliases, but couldn’t convince my family to make a switch :joy: I got roped into using Pass and Drive because it was there, but oversaw along the way that the Email offering is not as good as other options out there and they’re focusing on the wrong stuff that no one cares about like Wallet. This pressure from the memos to innovate confirms that. --- ## Post 23 by @trilobyte — 2026-11-01T13:28:16Z The CEO is the main shareholder, so I don’t see him going anywhere. Most Proton users will also not be aware of this or care. I think proton email is the best overall, but there’s other good options like Tuta. I have a thundermail beta account and may stay with them if they add PGP support and a lower price tier, otherwise I’ll go to Tuta who have stronger encryption than Proton and a good PWA. And yeah, [mailbox.org](http://mailbox.org) and [posteo.de](http://posteo.de) stay narrowly focused on email, are cheaper, and lack controversies, so they seem like a good choice. The security and usability will be a downgrade from Proton and Tuta though. Pass can be replaced by vaultwarden, keepass, or bitwarden easily. Drive I only use for sharing documents and photos. For backups, it’s not too hard to set up a home system. Proton docs is the one I can’t see giving up. --- ## Post 24 by @ph00lt0 — 2026-11-01T13:48:06Z We recommend tools based on being the best offering on privacy. I do not see how this would in anyway impact that right now. Most code contain AI generated contributions at this point including all the major operating systems and programming languages. We need to deal with the new reality find new ways to keep things secure, not swim against the stream. It won’t be a good strategy. While I also not not like the way that Andy Yen seems to get across to his staff, this is internal politics of the company that as far as I can tell are not at impact to the quality of products in such way it is worth this discussion. I feel like we are discussing drama here to be honest, aside from how bad the communication of Proton’s CEO here was. In fact I personally think it is a positive thing that Proton works with AI to defend against the new threats that are also caused by AI. I don’t like how AI has disrupted the world, but we have to live with it. AI doesn’t change any of the good and bad practices in software engineering, but it does massively increase timelines in both defense and offensive area. Knowing how to work with and against it, is important for the companies future existanceand your privacy. Proton Lumo (with apertus) at the very least is one of the more promising AI offerings that is not from the USA. --- ## Post 25 by @Sorge — 2026-11-01T13:49:22Z > [@trilobyte](#): > > I have a thundermail beta account and may stay with them if they add PGP support I never fully got it when people keep saying a service should add pgp support. If user wants pgp, they generates keys, publishes them, and that’s it, pgp is supported. That works for Tuta, Thundermail, anything else. Thunderbird supports PGP natively, and for Tuta Mailvelope works as well. It doesn’t cover the case where incoming mails are not encrypted, but that’s another story: tuta, posteo have better encryption anyway for such cases, idk about thundermail --- ## Post 26 by @trilobyte — 2026-11-01T14:44:59Z Stalwart allows the user to upload their key and then all incoming mails or stored mails are automatically encrypted in a way the admin can’t access. They just need to add a UI for users to upload their key to the stalwart server. --- ## Post 27 by @Boris — 2026-11-01T22:12:57Z > [@ph00lt0](#): > > We recommend tools based on being the best offering on privacy. Trust is also a factor, and when a company that is vocally anti-censorship in their marketing materials - applies censorship so liberally to their perceived benefit (as they have been systematically doing on the ProtonMail subreddit they control today) - it erodes that trust. Between Proton’s (future?) enshifitication, Tuta’s awkward UX and Mailbox’s legacy issues (anti-spoofing measures, TOTP implementation) - the bar for email is so low that maybe other players need to be added to the recommendations that are lacking in other ways. --- ## Post 28 by @Menkork — 2026-11-01T22:33:28Z > [@Boris](#): > > (as they have been systematically doing on the ProtonMail subreddit they control today) Not sure if I’d call whats currently going on censorship. That Proton snark subreddit showed up on my “Home” page (gee, thanks for that, Reddit), and even they were getting irritated by the endless flood of “I’m leaving” posts in their own sub lol. I can only imagine the sheer number of repetitive, low effort threads the moderators of the official subreddit have been dealing with the past few days. --- ## Post 29 by @Sorge — 2026-11-01T22:56:40Z Please think about how we use the word “censorship.” I can’t say “I want to see my photo on the main page of the Proton website, and if I don’t see it tomorrow, Proton supports censorship.” It’s their community, with their mods and policy. And on Reddit there’s another community about Proton products, without any Proton mods. I mean, you can trust anyone (I personally trust only @Linus_Sex_Tips), but you can’t put political labels on everything you don’t like. --- ## Post 30 by @Superman — 2026-11-02T03:02:49Z > [@Boris](#): > > Trust is also a factor, and when a company that is vocally anti-censorship in their marketing materials - applies censorship so liberally to their perceived benefit (as they have been systematically doing on the ProtonMail subreddit they control today) - it erodes that trust. Sigh. Man, you have no idea how often I see someone shout “censorship” when posts are delegated to a single thread. It’s literally a thing forums have been doing since the beginning. Hell, this forum does it. It’s literally rule 2 and on the protonmail subreddit. "_2 Avoid duplicate posts"_ Also, don’t forget this is reddit. With reddit mods. You know, like this [guy](https://www.youtube.com/watch?v=67CXV4o4dYM&t=0). No one pays reddit mods. They’re in it for the game!:person_facepalming: Is there stuff to criticize? Totally! But don’t just forgot how the internet works because you want to get mad. --- ## Post 31 by @dngray — 2026-11-02T08:42:30Z It’s worth noting while the comments weren’t exactly the most tactfully put, they were made within a closed environment to employees only as opposed to public through a corporate PR filter. We also don’t have anything more than the screenshots, and maybe there is other information that is important to the story. For example we don’t have access to performance data on employees, or how much they get paid. Maybe he’s annoyed because he can tell some employees are actually slacking off and not doing enough work for what they’re being paid? If that was true we wouldn’t know. I think it’s also important to know that thoughtful use of AI, isn’t a bad thing, and it’s not something that is going away. Humans do make mistakes, and get lazy as well. The latter is particularly obvious if you’ve ever had to review unit tests for example. --- ## Post 32 by @Aflame-Blighted — 2026-11-02T13:50:41Z > [@Boris](#): > > Between Proton’s (future?) enshifitication, Tuta’s awkward UX and Mailbox’s legacy issues (anti-spoofing measures, TOTP implementation) - the bar for email is so low that maybe other players need to be added to the recommendations that are lacking in other ways. Do you think Mozilla is too far gone for Thundermail to be included in the future? --- ## Post 33 by @object2598 — 2026-11-02T17:10:29Z Thundermail is based on Stalwart, there’s no particular encryption on it. I don’t see why would this be recommended at all. I’d rather self-host at this point. --- ## Post 34 by @QuickFox69 — 2026-11-02T22:00:54Z How do you get over the issues regarding your emails getting bounced by the big players? --- ## Post 35 by @Aflame-Blighted — 2026-11-02T22:39:51Z Ah. I see now that they have E2EE for their Send product, but not Mail. --- ## Post 36 by @trilobyte — 2026-11-02T22:55:47Z Currently it’s on the thundermail roadmap, but is a low priority with no stated plans about when it will be implemented: > **[Additional Encryption at Rest with OpenPGP or S/MIME | Thunderbird](https://ideas.tb.pro/p/additional-encryption-at-rest-with-openpgp-or-s-mime)** > > What you’d like to see Stalwart, the email server we are using, supports a feature where all incoming email is encrypted with the user’s OpenPGP or S/MIME key before it is stored. This is great because only the user can decrypt the email and it is... It says 2026 roadmap, but obviously we are approaching the end of 2026. We don’t know when they will get around to it. I can find no evidence they’ve started working on it or even that they have concrete plans about how it will work.