Proton Spam and the AI Consent Problem
Recorded: Jan. 23, 2026, noon
| Original | Summarized |
Proton Spam and the AI Consent Problem – David Bushell – Web Dev (UK) Homepage Contact Menu Home About Services Blog Notes Contact Proton Spam and the AI Consent Problem Subscribe Blog RSS feed Notes RSS feed Combined Mastodon Bluesky Thursday 22 Jan Play Synthesised Audio On Jan 14th Proton sent out an email newsletter with the subject line:Introducing Projects - Try Lumo’s powerful new feature nowAltscreenshot of the official email from @lumo.proton.meLumo is Proton’s “AI” offering.There is a problem with this email. And I’m not talking about the question of how exactly AI aligns with Proton’s core values of privacy and security.The problem is I had already explicitly opted out of Lumo emails.Altscreenshot of an unchecked toggleThat toggle for “Lumo product updates” is unchecked. Lumo is the only topic I’m not subscribed to. Proton has over a dozen newsletters, including some crypto nonsense. I opt-in to everything but Lumo, I gave an undeniable no to Lumo emails.So the email I received from Proton is spam, right?My understanding is that spam is a violation of GDPR and UK data protection laws. Regardless, Proton’s email is a clear abuse of their own service towards a paying business customer.Before I grab my pitchfork I emailed Proton support.Proton SupportDespite the subject line and contents, and despite the “From Lumo” name and @lumo.proton.me address, maybe this was an honest mistake?Proton’s first reply explained how to opt-out.Altscreenshot of support email quoted belowI’ve blurred the name because whateverHello David,Thank you for contacting us.You can unsubscribe from the newsletters if you do the following:- Log in to your account at https://account.protonvpn.com/login- Navigate to the Account category- Disable the check-marks under “Email subscriptions”- If you need additional assistance, let me know.[screenshot of the same opt-out toggle]-Have a nice day.John Support directs me to the exact same “Lumo product updates” toggle I had already unchecked. I replied explaining that I had already opted out. Support replies saying they’re “checking this with the team” then later replies again asking for screenshots.Can you make sure to send me a screenshot of this newsletter option disabled, as well as the date when the last message was sent to you regarding the Lumo offer?You can send me a screenshot of the whole message, including the date.Is it perhaps 14 January 2026 that you received the message?I found that last line curious, are they dealing with other unhappy customers? Maybe I’m reading too much into it.I sent the screenshots and signed off with “Don’t try to pretend this fits into another newsletter category.”After more “checking this with the team” I got a response today.In this case, the mentioned newsletter is for promoting Lumo Business Suit to Business-related plans.Hence, why you received it, as Product Updates and Email Subscription are two different things.In the subscription section, you will see the “Email Subscription” category, where you can disable the newsletter in order to avoid getting it in the future.If I understand correctly, Proton are claiming this email is the “Proton for Business newsletter”. Not the “Lumo product updates” newsletter.I don’t know about you, but I think that’s baloney. Proton Support had five full business days to come up with a better excuse. Please tell me, how can I have been any more explicit about opting out of Lumo emails, only to receive “Try Lumo” “From Lumo”, and be told that is not actually a Lumo email?Non-ConsentHas anyone else noticed that the AI industry can’t take “no” for an answer? AI is being force-fed into every corner of tech. It’s unfathomable to them that some of us aren’t interested.The entire AI industry is built upon a common principle of non-consent. They laugh in the face of IP and copyright law. AI bots DDoS websites and lie about user-agents. Can it get worse than the sickening actions of Grok? I dread to think.As Proton has demonstrated above, and Mozilla/Firefox recently too, the AI industry simply will not accept “no” as an answer. Some examples like spam are more trivial than others, but the growing trend is vile and disturbing.I do not want your AI.Update for 23rd JanuaryI guess someone at Microsoft read my post and said “hold my beer”. This morning I woke up to a lovely gift in my inbox; “Build Al agents with the new GitHub Copilot SDK”.AltGitHub email subject line: 'Build Al agents with the new GitHub Copilot SDK'GitHub Ensloppification is moving faster than I can delete my account for good. (It’s an unfortunate requirement for client projects.) For the record, I have never said “yes” to any GitHub newsletter. Even before Copilot I disabled every possible GitHub email notification.The “Unsubscribe” link provides the hidden newsletter list. There is nothing within GitHub account settings I can find to disable spam.AltGitHub 'Opt-Out Preferences' with 3 newsletters unchecked but GitHub Copilot emails checkedAs expected, Microsoft has opted me in without my consent. The wheels are falling off at GitHub. The brutally slow front-end UI. The embarrassingly lacklustre Actions CI. Now this sloppy tripe everywhere. Reminder to developers: GitHub is not Git. Subscribe Blog RSS feed Notes RSS feed Combined Mastodon Bluesky David Bushell I design, build, and consult upon all things web. Based in the UK and coding worldwide. Hire me! About Mastodon Bluesky CodePen Forgejo RSS Feed My Projects New and old! Read my blog and notes, see what I’m coding now and my design portfolio. Web Agencies Need a hand? I’m reliable, flexible, and pride myself in communication and initiative to delivery on time and within budget. Businesses Have an idea? Let’s discuss your requirements, share ideas, and figure out what’s best for your new website together. The Blog Search for Search Hmmarkdown 2 Proton Spam and the AI Consent Problem Winter 2026: Self-Hosted… Success! Winter 2026: Self-Hosted… Disaster? Death to Scroll Fade! App Defaults 2026 I Survived 2025 Wrapping up Advent of Code 2025 Deleting Code for Performance iOS Web Browser Latency Subscribe Blog RSS feed Notes RSS feed Combined Mastodon Bluesky Copyright and Privacy Copyright © 2004–2026 • David Bushell • Privacy Policy • AI Policy Settings Site Settings Close Colour scheme |
David Bushell’s recent blog post, “Proton Spam and the AI Consent Problem,” highlights a troubling trend within the rapidly expanding artificial intelligence industry: the persistent disregard for user consent. The core of Bushell’s argument centers on an interaction with Proton, specifically regarding their “Lumo” AI offering and the subsequent emails he received despite explicitly opting out. Bushell received an email promoting Lumo, identified with the “From Lumo” address, despite having already deactivated the “Lumo product updates” subscription within his Proton account. Proton’s initial response focused on directing him back to the same toggle he’d previously used to unsubscribe, indicating a lack of genuine understanding of his stated preferences. The support team’s insistence that the email was part of the “Proton for Business newsletter,” rather than the unwanted Lumo promotion, felt deliberately obfuscating, and highlighted a broader manipulation tactic. Bushell’s experience echoes a wider concern regarding the AI industry’s approach to user interaction. The relentless integration of AI into various platforms—from email newsletters to code completion tools—suggests a broader pattern of disregarding explicit “no” responses. This isn't limited to isolated incidents; the author highlights further examples involving Microsoft’s GitHub Copilot aggressively offering updates, despite his diligent efforts to disable all related notifications. The case against Proton, while seemingly minor on the surface, represents a symptom of a larger problem. It demonstrates a failure of communication and a lack of respect for individual user choices within an industry increasingly driven by data collection and engagement, regardless of explicit opt-outs. Bushell argues that the AI sector is fundamentally built upon a principle of non-consent, yet this is routinely ignored. The response from Proton, characterized by redirection and obfuscation, further underlines this issue. The entire situation serves as a cautionary tale, indicative of a concerning trend where user agency is increasingly disregarded in the pursuit of AI adoption. Ultimately, Bushell’s post raises important questions about ethical considerations—specifically regarding consent and the ongoing influence of AI within the digital landscape. |