Google must have some awful PMs and designers. The worst UX decision I have seen recently is AI auto-dubbing all youtube videos by default with no way to disable this behavior globally. How could you miss that people can be fluent in multiple languages and if I click on a video in a foreign language, I most likely want the original soundtrack? Clearly, the intention was to boost some metric “X users are using this feature” with no regard for the actual user.
> The worst UX decision I have seen recently is AI auto-dubbing all youtube videos by default with no way to disable this behavior globally.
Recently anyways. The most egregious thing about Youtube, which is not terribly new, is the Shorts. If your video is short enough, it is auto-converted to a "Short", and the original aspect ratio gets cropped to be vertical orientation (for viewing on a phone, presumably).
It's wild how disrespectful this is to viewers and content creators. Google literally couldn't care less what people want, it's just a machine that optimizes for whatever KPIs a given manager has that quarter, users be damned.
Of course Google doesn't have to care about users because they have a dozen different monopolies. It's sad that they were allowed to get to this point.
The most irritating thing for me now is that YouTube doesn’t work in my browser anymore. Clearly being a/b tested because sometimes it does, and sometimes it spews out thousands of console errors and doesn’t load anything.
Yes. Presently, with ublock origin on fairly default settings, YouTube causes an unresponsive tab that doesn't play the video, and a blast of errors on the console. Disabling UBO on youtube.com fixed the problem instantly for me (which is fine since I pay for YouTube Premium because it's the correct solution to the problem of "pay content creators and don't destroy my experience")
I watch about 50% Japanese content and having to switch this off manually has become a major source of annoyance. Bizarrely, if I misunderstand something in Japanese and want to go back to check, these same videos normally don't have any English subtitles available. There's auto-transcribed Japanese subtitles which are about 90% accurate, but they're rarely translated.
The absolutely wild disparity in compute required to translate the Japanese text to English vs rendering an entirely new soundtrack in English blows my mind. I guess someone at Google thought it made sense because many people prefer dubs to subs, but that's on highly polished entertainment product vs 1-person-and-their-Japanese-vlog channels which are not aiming at a mass audience.
if it increases topline metrics like watch time it's probably hard for them to justify removing it. a change this big seems like it was probably a/b tested and did move metrics significantly?
Probably. We keep watching all kind of stuff after getting baited into it. AI slob is annoying, but we do want to know what chefs do about sticky pizza dough, or what that secret in the pyramids is, or how the kid reacted to what the cat did, or (insert your guilty pleasure here).
on some platforms I try to be really good about hitting the "Never recommend this channel/page/whatever again" whenever the algo serves me the bottom-tier gutter trash videos, such as the "idiotic life hack that obviously won't work" engagement bait. It's a small drop in the ocean, but at least that one channel will never be served to me again.
I complained about the same problem a while ago. There are a few recommendations for alternative YT interfaces that don't seem to be as messed up yet: https://news.ycombinator.com/item?id=45503218
And apropos of nothing... there's another link on the front page at the moment ( https://news.ycombinator.com/item?id=45850430 ) talking about Apple crossing the red line of customer satisfaction. This got me thinking... to Alphabet, you're not the customer, you're the product. YouTube is a sugary trap to lure eyeballs into the advosphere.
YouTube is one of the better ones -- sort of -- in that you can simply pay with a very small amount of money rather than by wasting a ton of time on ads.
Only "sort of" though since they still use the same spammy algorithm-driven timeline and Shorts and stuff, and are clearly still trying to maximize your total watch time. Given that I just pay a fixed fee, I wish they'd use a different algorithm that only seeks to keep my engagement with YouTube from dipping too low, rather than the default which is clearly designed to turn a 23-hour-a-day user into 24.
Go to your Google account settings; add the languages you speak and don’t want auto translations for in your personal profile.
I agree that the auto dubbing is the worst feature. It may have been HN where I read the above tip to turn that off, it seems to have worked for me so far.
In addition to the feature being auto-on (for me, at least) and unasked-for, you also need to perform multiple clicks through non-obvious menus (I think one of them was "Audio track"?) to get to the original audio. Another layer of obnoxiousness.
Yeah, they do. It's just that the customers have changed. Once they were the viewers, then the advertisers, and now that monopoly power is secure, the "customers" of those PMs are the financial backers.
This happened to me for the first (and only time so far) the other day on a video that wasn’t even in another language but from an guy who sometimes posts in another language (but usually in English) but does have a strong German (I think) accent. I was so confused at first and it took me a while to figure out what was going on as I could tell his voice was weird and then noticed the audio was completely out of sync.
Reddit has machine-translated their whole site in different languages and made these translated versions google-indexed. That polluted google result severely. As if google weren't bad enough today.
True but that is a function of ignorance too. There are plenty of good features in Gmail that are off by default, like undo / delayed send and keyboard shortcuts.
"Enable delay send - Allows you to undo sending emails for 5 minutes", I'd argue that a lot of people would enable that pretty fast.
Keyboard shortcuts probably would work like I'd expect, people like me would go "Hell no, no keyboard shortcuts in browser application EVER", and power users would opt into that in an instant.
I'm confused. Doesn't gmail offer an 'undo' for send by default ? At least for the last 5 years ? It's in General settings "Undo Send" and can be set up to 30 seconds ?
I really wished they would also let you disable those very annoying modal popups announcing yet-another-chatbot-integration twice a week: My company is already paying for your product, just let me do my work ffs...
And this is generically true and always has been about every aspect of GMail?
What would you suggest people do. Self-host?
I'm just trying to understand why you posted this. It's generically true. Any company can change anything at any point. May as well just pack it up boys.
> I'm just trying to understand why you posted this. It's generically true. Any company can change anything at any point. May as well just pack it up boys.
Yes, any SaaS can change any feature at any time. Some companies have different motives though. We're not paying for GMail. When customers pay a monthly subscription and can cancel at any time, you usually want to keep them happy.
The internal motives are also different. Are employees promoted for just launching stuff? Are they running out of helpful features to launch?
> And this is generically true and always has been about every aspect of GMail?
In principle, but look at all the ways Gmail bends over backwards to keep ancient UI preferences working. You can configure it for different inbox presentations, different densities, snippets or not, images displayed or not, UI icons or text, you can disable and enable threading, you can put chat and meet on one side or not, you can have keyboard shortcuts or not, you can remap all the keyboard shortcuts if you use them, etc etc etc.
You can't self-host these days -- your emails will get stuck in every kind of spam filter there is, and you'll always have cause to wonder if your emails are received by their intended recipients, or lost to the abyss.
You've got to use either Gmail, Microsoft, Protonmail, etc. I don't love them, but Proton is probably the best of a bad bunch.
> You can't self-host these days -- your emails will get stuck in every kind of spam filter there is, and you'll always have cause to wonder if your emails are received by their intended recipients, or lost to the abyss.
this is not true unless you end up on an IP previously abused
if you don't want to take on the risk at all, there's email services for pennies / thousand emails
Well, they're off by default in the countries mentioned in the top-level comment because they're legally required to be opt-in there (the implementation rather than the feature of course, but it couldn't really be otherwise).
I suppose, to your point, Google doesn't have to make it optional in other countries... But that discrepancy would seem to have a lot of downside (maintenance, optics, docs) for little upside (...force adoption against the will of users who would go out of their way to opt out of they could?).
I would really love it if there was a "smart setting" (or a dumb setting) to prevent people from sending me their drip marketing spam. The spam filtering in my personal Gmail is adequate, if not perfect (I really don't understand how the constant life insurance spam is getting through). But my main client uses Google Workspace or GSuite or whatever it is called these days, and my inbox for that email features a constant barrage of drip campaign garbage.
Are Google's incentives misaligned in some way here? It's not like the heuristics are particularly difficult for this kind of email. Some of it even has unsubscribe links (I didn't subscribe), or, "If you don't want to hear from me again, just let me know", etc.
Some of it even has unsubscribe links (I didn't subscribe), or, "If you don't want to hear from me again, just let me know", etc.
I don't know if this counts as "drip marketing" (a new term to me), but just this week Apple spammed me with some Apple Card offer for Hertz car rentals.
No way to unsubscribe. No link. No mention of unsubscribing at all. And on the Apple Card web site, no way to turn off marketing emails.
I wish you could still report these spam's to the FTC.
I run my email via imap and haven't seen GMail web UI for at least 15 years. Apple does some minor changes to Mail app but generally they follow "if it ain't broke don't fix it" motto, and I really appreciate that. Besides, I have email from Google, Yahoo, Microsoft, and FastMail in the same app, and I really appreciate that all email for me looks the same.
Good for you. 9999 out of 10,000 people don't know what a "gmail web UI" is, or even that there's something called email that is separate from the gmail web interface.
True, and 10-20 years ago this seemed like a transitional issue: the world of technology was new and humanity needed time to adapt. Today I feel like some of this should be taught at school. Stuff like what is a browser, what is TLS, what are cookies, what is an email, what is phishing, etc. I know schools used to teach people Excel, Word and other office programs. Maybe they still do, and web should be a part of that curriculum.
Pretty sure most people access their mail on their phone or tablet these days, either using a service-specific app, or a generic app like Apple Mail. I wouldn't be surprised in the least if many Gmail users had never used the web interface at all.
The above poster is not talking to end users about the web interface. The poster is talking to you. About the fact that end users don't separate the interface from the thing.
You are making the same argument with the same language to someone that made the same point.
I've never flagged a submission before, but I just have to with this one.
There's no evidence, no blurred-out screen shot even, no one here has seen this behavior in the comments, and I can't find any corroborating reports on the web.
Anyone remember the news story about Metallica copyrighting G and F chords? It’s sort of like that, true or not it’s so on brand that people just accept it, it doesn’t tell you something you did already know about how google (or Metallica) behave.
I just flagged yours because I don't think it makes any sense to flag the post much less posting about flagging it - especially when you said you are an ex-googler.
Life continues, move on, defending Google enshitfication online after being laid off is extra work for no payment, you can rest.
100% Google has been making their AI more intrusive and in your face across all their portfolio. It's not just Google, Atlassian is doing the same
With search in gcloud, the drop-down top 2/3 is ai calls to action. Completely useless because their suggestions are so bad and for such basic tasks that I never do.
It feels like in platform advertising.
I've left them feedback, and since they've only doubled down, am now reducing my spend
GMail started inserting package delivery notifications at the top of my inbox screen. Not exactly ideal for Christmas shopping with family milling about. If you turn it off, you lose the tabs for email categorization, so instead I wrote a CSS rule to hide it permanently, but it's a bad feature.
am i missing something here, or is this really just a random two-sentence gripe from some guy, complaining about something that nobody else can reproduce?
am i missing something here, or is this really just a random two-sentence gripe from some random guy, complaining about something that nobody else can reproduce?
You have just summarized HN, and most of social media and the blogosphere, in 2025.
The internet is 44% angry people getting off on being angry about nothing.
Tangentially related, the AI integrated in Google Chat is hilariously bad. Find a thread which starts with „Bug: (…)“ that has 90+ answers. Hey, an AI could be useful here! Click summarize. Wait. Without fail the result will be along the lines of „X, Y and Z discuss a bug.“
It would be fine if you could then reply with 'no, please tell me about the nature of the bug not the people involved', and then have it remember that forever.
However nearly nobody seems to correctly implement this user-wide memory.
I've been a long-time Gmail user with a paid plan. But recently two issues are annoying me enough that I'm considering leaving:
1. The incessant "Using Gmail to run your business?" upsells. No, I'm unemployed and this is a personal account. Unlike the AI upsell, I can't seem to dismiss this permanently. It just snoozes it until the next time I open it.
2. The Search bar has become dangerously glitchy (at least in Firefox for Mac) if you type fast and have keyboard shortcuts on. It lets me type 1 or 2 characters before it starts treating every character as a shortcut, inadvertently deleting, muting, archiving emails. Search is what sets Gmail apart and now it is unusable. I reported this bug to Google months ago and my patience is running thin.
I paid for and tried Google AI for Gmail and was appalled at how bad it is. The product team there is really not executing well. I've switched now to Shortwave. It works very well and having LLM+RAG queries for 20+ years of email archives has been very helpful to me.
Something I noticed recently is that titles of youtube videos are automatically translated.
I'm convinced that nobody in Google speaks more than one language, otherwise they would have never done that. It is impossible to turn off the unsolicited translations in search, and now youtube. I'm scared that soon they will force up audio translation as well.
All this AI is great. What I really want is to clean up my inbox by seeing where the mails are coming from. What I want is `select sender, count(*) from emails group by 1 order by 2 desc` or the equivalent. But I guess that doesn’t juice the stock price like AI does.
I can see how it is useful: for some people, reading and writing emails is a dreadful activity. But it is also an important facet of every human activity -- communication. So, it better be intentional and clear. I have turned off all smart features and let me handle how to say what I want to say.
Man I was in a Google News Initiative meeting and they started with a "fun poll" where they asked how you'd feel if you read an email that you knew was written by AI. Lots of people didn't mind knowing that was the case, and it was mainly for the reasons you put out writing emails is legit not fun for a lot of people.
Man I was so triggered though, like old-man-yelling-at-clouds level. Then I went back through gmail and exactly one of the last 50 most recent emails I've received was not an alert or a newsletter or an invite. IOW out of my last 50 emails only one of them would've fit the bill for an AI-written correspondence taking the place of a human-written one. All that butt-hurt and it doesn't even matter.
So long way of getting around to saying that your point about intentionality and clarity is clutch, and if people can get that with AI-written emails maybe I can cool my jets.
One of my tech clients emailed me about a Roku problem and one of the AI suggestions in Gmail was to tell the customer to try unplugging a Roku for a minute and plugging it back in. I pressed that button, proofread the suggested text, and hit send.
In truth it felt both amazing and made me uneasy, for AI was encroaching on my career of telling people to reboot their errant device.
i do about 95% of my emailing in thunderbird on my computer, and 5% in the gmail app on my phone or on the specific email's website. all imap so they're basically synced already. i'm wondering if your marco app will be useful for my use case, and what the benefits would be. i think i would lose some benefits such as thunderbird extensions. also, to use proton mail on my phone i think i need the protonmail app. i use the proton bridge to use it in thunderbird. would the marco app work with proton on computer but not on phone?
Is there a way to suspend/disable Gmail account in a way, that no bad actors could overtake it, after Google disables/deletes(?) it after period of two years?
I've seen no indications of this supposed feature in my own experience or other sources. The article does not provide any evidence to support its claims.
It's amazon deliveries for me. Gmail gets confused about multi-item orders with different delivery dates, and I cannot click away the banners displaying on top of the mailbox. Really poor UI.
OTOH, it's nice to see that there is some innovation happening - eventually (I am an optimist) they will weed out the lousy stuff and keep the useful.
What makes you think it hasn't long been using a large language model for it though? Maybe not Large Language Model, the term seems to have come with ChatGPT et al., but Google's been doing machine learning for a long long time; the transformer architecture paper (kind of the groundwork for the current wave) was published by a Google team in 2017.
A lot of new Google features are branded 'AI', because it's so hyped it has broad consumer awareness, but a lot of Google features for a long time have used AI and just been brandless or at least 'AI'-less features.
It started about a month ago for me. I am subscribed to emails from an IT as a Gig platform called Field Nation. Thought I might pickup a bit of side work but never did.
Recently it started adding them to my calendar and there is no way to turn off this feature without also turning off useful features such as package out for delivery notifications.
Where I work, we get tons of feedback from customers, basically saying, "How do I disable this new AI feature? I never asked for this." Yet, the guidance from leadership is very clear: "More AI!" I suspect it's like this at Google (and everywhere else), too.
Yeah, you gotta watch out for this stuff. The other day I was using Gmail when it just gave me a million dollars. I gave it to charity under an assumed name.
Can’t screenshot without revealing the name but I think it’s actually super cool of Google to do this. Just distributed charity. With no benefit to themselves.
They did say that they did it based on my Google Photos showing that I was really good looking (can’t share because of privacy) so there’s some privacy stuff there but overall I think it’s good.
Did the user accidentally keep hitting Alt+H? Or are they part of some experiment? Gmail writes nothing for me unless I click the "help me write" link. I also don't know why it is so hard to record a screenshot of the behavior, you can write it to example@example.com, and the draft email UI can be clipped from where it shows your email address.
> Gmail writes nothing for me unless I click the "help me write" link.
For me it sometimes (I'm not sure why, it isn't even "intelligent" about which emails have options, automatic noreply@ emails sometimes have it) has buttons to pre-fill a reply with a message out of a few choices. It's not unprompted, but I could see someone accidentally clicking it instead of the reply button.
I see obviously AI-powered tab-completion suggestions, and the Alt-H prompt suggestion when first writing an email, but I've never had it actually insert text unprompted.
And yet google's spam filters still let through so much AI slop spam. Any email from a new sender should be suspect, and if it has an awkwardly phrased opt-out line ("if you'd rather not hear from me again, just shoot me a "leave out"") then it should presumptively be marked as spam.
I have the gmail app set to only notify me for "important messages," and it pretty reliably doesn't trigger a notification for this junk. I wish I could have it use that system (whatever it's using) to actually flag messages as spam/mass emails.
If it's a seemingly personal message from an unknown sender, send it to spam. If it looks like a mass email (that someone might have signed up for), then put it in the Commercial tab or whatever it's called, so I can set up a filter for it or unsubscribe.
I think letting your email app raise notifications is a mistake, and I can see how letting it do that would amplify the annoyance of any misclassified messages.
Until about 9 months ago it was not an issue. I'm glad the gmail app has the "important only" feature, but just wish it was given more power so that when I'm at my computer they are properly categorized there as well.
Please do yourself and the world a measure of good, and stop using GMail. Giving someone a GMail address is like saying "Yes, I like to be abused, I like to be violated and have no privacy". It is quite embarrassing - and people who are tech-savvy should be telling their colleagues and peers that it is embarrassing and inappropriate.
Yes, I realize that dropping the GMail mailbox does not mean you are free from GMail, because of all of the other people on it. But we must each make some effort to chip away at that thing. Twenty years ago, when Microsoft's hold on PCs was much stronger than it is today, online venues were inundated with derision and denouncement of them and their practices; but these days - it's as though Alphabet/Google control of all these aspects of so many people's lives is just neutral reality, ho-hum, nothing to waste time thinking about.
I moved to ProtonMail right before their CEO started mouthing off about his dipshit opinions. I still use them because it's cheaper to pay for a family plan and get a VPN, password manager, and storage along with the email. I used to pay for each of these services individually and was paying 2x what I pay now
> I used to pay for each of these services individually and was paying 2x what I pay now
What's the other factor though and is it worth the getting walled in?
Password manager in my opinion has to be its own thing, has nothing to do with the rest, same for VPN. I can see email, calendar and data being close together as being beneficial in every day flow. That's where Fastmail for instance draws the line I think.
If you untangle these, would the cost not be worth the benefit?
ProtonMail also does this. Someone actually told me they got an email from me and when I checked it said “this mail was AI generated”. I can’t screenshot it though. Personal reasons.
Google must have some awful PMs and designers. The worst UX decision I have seen recently is AI auto-dubbing all youtube videos by default with no way to disable this behavior globally. How could you miss that people can be fluent in multiple languages and if I click on a video in a foreign language, I most likely want the original soundtrack? Clearly, the intention was to boost some metric “X users are using this feature” with no regard for the actual user.
> The worst UX decision I have seen recently is AI auto-dubbing all youtube videos by default with no way to disable this behavior globally.
Recently anyways. The most egregious thing about Youtube, which is not terribly new, is the Shorts. If your video is short enough, it is auto-converted to a "Short", and the original aspect ratio gets cropped to be vertical orientation (for viewing on a phone, presumably).
It's wild how disrespectful this is to viewers and content creators. Google literally couldn't care less what people want, it's just a machine that optimizes for whatever KPIs a given manager has that quarter, users be damned.
Of course Google doesn't have to care about users because they have a dozen different monopolies. It's sad that they were allowed to get to this point.
I tend to raise holy hell about Whatsapp because that's pretty clear antitrust and Zuckerberg lied
However with that also being said Google is a menace
That is extraordinarily hostile to both creators and consumers.
Gotta pump those engagement metrics for shorts.
The most irritating thing for me now is that YouTube doesn’t work in my browser anymore. Clearly being a/b tested because sometimes it does, and sometimes it spews out thousands of console errors and doesn’t load anything.
Yes. Presently, with ublock origin on fairly default settings, YouTube causes an unresponsive tab that doesn't play the video, and a blast of errors on the console. Disabling UBO on youtube.com fixed the problem instantly for me (which is fine since I pay for YouTube Premium because it's the correct solution to the problem of "pay content creators and don't destroy my experience")
You don't have any old browser extensions by any chance? "Enhancer for YouTube" is an extension which has become unmaintained, and will break YouTube.
No I have a few privacy plugins/adblock but they all work fine with YouTube normally. I tried disabling everything. All the standard debugging tests.
I watch about 50% Japanese content and having to switch this off manually has become a major source of annoyance. Bizarrely, if I misunderstand something in Japanese and want to go back to check, these same videos normally don't have any English subtitles available. There's auto-transcribed Japanese subtitles which are about 90% accurate, but they're rarely translated.
The absolutely wild disparity in compute required to translate the Japanese text to English vs rendering an entirely new soundtrack in English blows my mind. I guess someone at Google thought it made sense because many people prefer dubs to subs, but that's on highly polished entertainment product vs 1-person-and-their-Japanese-vlog channels which are not aiming at a mass audience.
+1, this is the most annoying YouTube feature I've ever come across. Gave them feedback on it.. maybe more people should complain
if it increases topline metrics like watch time it's probably hard for them to justify removing it. a change this big seems like it was probably a/b tested and did move metrics significantly?
Probably. We keep watching all kind of stuff after getting baited into it. AI slob is annoying, but we do want to know what chefs do about sticky pizza dough, or what that secret in the pyramids is, or how the kid reacted to what the cat did, or (insert your guilty pleasure here).
on some platforms I try to be really good about hitting the "Never recommend this channel/page/whatever again" whenever the algo serves me the bottom-tier gutter trash videos, such as the "idiotic life hack that obviously won't work" engagement bait. It's a small drop in the ocean, but at least that one channel will never be served to me again.
I complained about the same problem a while ago. There are a few recommendations for alternative YT interfaces that don't seem to be as messed up yet: https://news.ycombinator.com/item?id=45503218
And apropos of nothing... there's another link on the front page at the moment ( https://news.ycombinator.com/item?id=45850430 ) talking about Apple crossing the red line of customer satisfaction. This got me thinking... to Alphabet, you're not the customer, you're the product. YouTube is a sugary trap to lure eyeballs into the advosphere.
YouTube is one of the better ones -- sort of -- in that you can simply pay with a very small amount of money rather than by wasting a ton of time on ads.
Only "sort of" though since they still use the same spammy algorithm-driven timeline and Shorts and stuff, and are clearly still trying to maximize your total watch time. Given that I just pay a fixed fee, I wish they'd use a different algorithm that only seeks to keep my engagement with YouTube from dipping too low, rather than the default which is clearly designed to turn a 23-hour-a-day user into 24.
Go to your Google account settings; add the languages you speak and don’t want auto translations for in your personal profile.
I agree that the auto dubbing is the worst feature. It may have been HN where I read the above tip to turn that off, it seems to have worked for me so far.
In addition to the feature being auto-on (for me, at least) and unasked-for, you also need to perform multiple clicks through non-obvious menus (I think one of them was "Audio track"?) to get to the original audio. Another layer of obnoxiousness.
I don't even know how to get the original video title back…
You can't. The translated titles and descriptions are ruined and there is nothing you as a viewer can do to fix it back.
I don’t think it’s just google. Modern product management feels broken across the whole industry. Do these guys even talk to customers anymore?
I think this is where “ you’re the product, advertisers are the customers” comes in.
Yeah, they do. It's just that the customers have changed. Once they were the viewers, then the advertisers, and now that monopoly power is secure, the "customers" of those PMs are the financial backers.
This happened to me for the first (and only time so far) the other day on a video that wasn’t even in another language but from an guy who sometimes posts in another language (but usually in English) but does have a strong German (I think) accent. I was so confused at first and it took me a while to figure out what was going on as I could tell his voice was weird and then noticed the audio was completely out of sync.
Quite happy with an extension which disables the feature (not happy with Youtube forcing the feature). https://github.com/YouG-o/YouTube-No-Translation
Reddit does the same for comment. It's on by default and the quality of translations isn't good. Completely jarring
Reddit has machine-translated their whole site in different languages and made these translated versions google-indexed. That polluted google result severely. As if google weren't bad enough today.
I just experienced this and was wondering why it was happening!
I assume all UX designers are replaced by marketing.
Anecdotally, I don’t see any of this. I have all “smart” features in gmail turned off; there is an option like this in the settings.
Google’s Help: https://support.google.com/mail/answer/15604322
Also relevant:
> By default, smart feature settings are off if you live in: The European Economic Area, Japan, Switzerland, United Kingdom
This. You can disable all smart features (which includes things like mail categories, AI auto-complete, and most things that look at your emails).
Gear -> All Settings -> General tab (default) -> Smart features: Turn on smart features in Gmail, Chat, and Meet
Linked help page: https://support.google.com/mail/answer/15604322
How many would turn them on if they defaulted to "Off"? Probably not enough to justify the development cost.
True but that is a function of ignorance too. There are plenty of good features in Gmail that are off by default, like undo / delayed send and keyboard shortcuts.
"Enable delay send - Allows you to undo sending emails for 5 minutes", I'd argue that a lot of people would enable that pretty fast.
Keyboard shortcuts probably would work like I'd expect, people like me would go "Hell no, no keyboard shortcuts in browser application EVER", and power users would opt into that in an instant.
Allow you to undo sending for 5 minutes means email delivery is delayed by 5 minutes.
It doesn't have to be 5 minutes. It could be 15 seconds. I've used that feature (in Fastmail), and find it very valuable.
I'm confused. Doesn't gmail offer an 'undo' for send by default ? At least for the last 5 years ? It's in General settings "Undo Send" and can be set up to 30 seconds ?
I really wished they would also let you disable those very annoying modal popups announcing yet-another-chatbot-integration twice a week: My company is already paying for your product, just let me do my work ffs...
> You can disable all smart features
For how long?
You don't own the platform. Google PMs may decide to roll it out to everyone at some point to hit numbers.
And this is generically true and always has been about every aspect of GMail?
What would you suggest people do. Self-host?
I'm just trying to understand why you posted this. It's generically true. Any company can change anything at any point. May as well just pack it up boys.
> I'm just trying to understand why you posted this. It's generically true. Any company can change anything at any point. May as well just pack it up boys.
Yes, any SaaS can change any feature at any time. Some companies have different motives though. We're not paying for GMail. When customers pay a monthly subscription and can cancel at any time, you usually want to keep them happy.
The internal motives are also different. Are employees promoted for just launching stuff? Are they running out of helpful features to launch?
> And this is generically true and always has been about every aspect of GMail?
In principle, but look at all the ways Gmail bends over backwards to keep ancient UI preferences working. You can configure it for different inbox presentations, different densities, snippets or not, images displayed or not, UI icons or text, you can disable and enable threading, you can put chat and meet on one side or not, you can have keyboard shortcuts or not, you can remap all the keyboard shortcuts if you use them, etc etc etc.
The point is to say that it’s bad and Google specifically can’t be trusted. It’s good to express disapproval of unethical business practices.
You can't self-host these days -- your emails will get stuck in every kind of spam filter there is, and you'll always have cause to wonder if your emails are received by their intended recipients, or lost to the abyss.
You've got to use either Gmail, Microsoft, Protonmail, etc. I don't love them, but Proton is probably the best of a bad bunch.
> You can't self-host these days -- your emails will get stuck in every kind of spam filter there is, and you'll always have cause to wonder if your emails are received by their intended recipients, or lost to the abyss.
this is not true unless you end up on an IP previously abused
if you don't want to take on the risk at all, there's email services for pennies / thousand emails
> if you don't want to take on the risk at all, there's email services for pennies / thousand emails
I'm seriously interested. Which ones would you recommend? Are they reliable?
Migadu, Fastmail, Protonmail, Zoho, Tutanota
These have all been running for many years and work fine, hell there's even the meme addresses at cock.li which has been running for over 10 years.
You don't need to be on a gmail account for reliable email.
Well, they're off by default in the countries mentioned in the top-level comment because they're legally required to be opt-in there (the implementation rather than the feature of course, but it couldn't really be otherwise).
I suppose, to your point, Google doesn't have to make it optional in other countries... But that discrepancy would seem to have a lot of downside (maintenance, optics, docs) for little upside (...force adoption against the will of users who would go out of their way to opt out of they could?).
This is often the case with Google products because the A/B testing is rampant.
I would really love it if there was a "smart setting" (or a dumb setting) to prevent people from sending me their drip marketing spam. The spam filtering in my personal Gmail is adequate, if not perfect (I really don't understand how the constant life insurance spam is getting through). But my main client uses Google Workspace or GSuite or whatever it is called these days, and my inbox for that email features a constant barrage of drip campaign garbage.
Are Google's incentives misaligned in some way here? It's not like the heuristics are particularly difficult for this kind of email. Some of it even has unsubscribe links (I didn't subscribe), or, "If you don't want to hear from me again, just let me know", etc.
Some of it even has unsubscribe links (I didn't subscribe), or, "If you don't want to hear from me again, just let me know", etc.
I don't know if this counts as "drip marketing" (a new term to me), but just this week Apple spammed me with some Apple Card offer for Hertz car rentals.
No way to unsubscribe. No link. No mention of unsubscribing at all. And on the Apple Card web site, no way to turn off marketing emails.
I wish you could still report these spam's to the FTC.
I run my email via imap and haven't seen GMail web UI for at least 15 years. Apple does some minor changes to Mail app but generally they follow "if it ain't broke don't fix it" motto, and I really appreciate that. Besides, I have email from Google, Yahoo, Microsoft, and FastMail in the same app, and I really appreciate that all email for me looks the same.
Good for you. 9999 out of 10,000 people don't know what a "gmail web UI" is, or even that there's something called email that is separate from the gmail web interface.
True, and 10-20 years ago this seemed like a transitional issue: the world of technology was new and humanity needed time to adapt. Today I feel like some of this should be taught at school. Stuff like what is a browser, what is TLS, what are cookies, what is an email, what is phishing, etc. I know schools used to teach people Excel, Word and other office programs. Maybe they still do, and web should be a part of that curriculum.
Anecdotally, most of the non-tech-savvy people in my life use gmail via Apple's Mail app on their phone or iPad.
Same. I know so many people who just use Apple Mail and don’t even know that there is or why they would use the gmail app.
Pretty sure most people access their mail on their phone or tablet these days, either using a service-specific app, or a generic app like Apple Mail. I wouldn't be surprised in the least if many Gmail users had never used the web interface at all.
> web interface.
Bold of you to think they know what is this.
I have seen enough people who wholeheartedly though the things in their phone stays in it and you lose access to them if you move to another phone.
You didn't read the post you responded to.
The above poster is not talking to end users about the web interface. The poster is talking to you. About the fact that end users don't separate the interface from the thing.
You are making the same argument with the same language to someone that made the same point.
I turned on Fastmail and all the AI bullshit went away for good. No regrets.
I've never flagged a submission before, but I just have to with this one.
There's no evidence, no blurred-out screen shot even, no one here has seen this behavior in the comments, and I can't find any corroborating reports on the web.
Who knows what's really happening or not here?
Anyone remember the news story about Metallica copyrighting G and F chords? It’s sort of like that, true or not it’s so on brand that people just accept it, it doesn’t tell you something you did already know about how google (or Metallica) behave.
https://www.metalunderground.com/news/details.cfm?newsid=575...
https://www.smbc-comics.com/comics/1505312920-20170913%20(1)...
Isn't this HN in a nutshell these days?
I don’t get it
Perfect response lol. Rare to see it's not xkcd.
I just flagged yours because I don't think it makes any sense to flag the post much less posting about flagging it - especially when you said you are an ex-googler.
Life continues, move on, defending Google enshitfication online after being laid off is extra work for no payment, you can rest.
100% Google has been making their AI more intrusive and in your face across all their portfolio. It's not just Google, Atlassian is doing the same
With search in gcloud, the drop-down top 2/3 is ai calls to action. Completely useless because their suggestions are so bad and for such basic tasks that I never do.
It feels like in platform advertising.
I've left them feedback, and since they've only doubled down, am now reducing my spend
Moving to Cloudflare, if you're curious
Cloudflare has a gmail competitor?
Not really 1:1 but I did find this
https://www.cloudflare.com/developer-platform/products/email...
I hate the “AI mode”
GMail started inserting package delivery notifications at the top of my inbox screen. Not exactly ideal for Christmas shopping with family milling about. If you turn it off, you lose the tabs for email categorization, so instead I wrote a CSS rule to hide it permanently, but it's a bad feature.
Not sure what you mean here. You can disable package status and still keep tabbed inbox.
am i missing something here, or is this really just a random two-sentence gripe from some guy, complaining about something that nobody else can reproduce?
"dave is mad at gmail". okay dave.
am i missing something here, or is this really just a random two-sentence gripe from some random guy, complaining about something that nobody else can reproduce?
You have just summarized HN, and most of social media and the blogosphere, in 2025.
The internet is 44% angry people getting off on being angry about nothing.
i feel like HN is usually at least a little bit better than this.
the comments are often full of silly gripes, but they don't usually get voted up to the top of the homepage and stay there for hours
Tangentially related, the AI integrated in Google Chat is hilariously bad. Find a thread which starts with „Bug: (…)“ that has 90+ answers. Hey, an AI could be useful here! Click summarize. Wait. Without fail the result will be along the lines of „X, Y and Z discuss a bug.“
It would be fine if you could then reply with 'no, please tell me about the nature of the bug not the people involved', and then have it remember that forever.
However nearly nobody seems to correctly implement this user-wide memory.
I've been a long-time Gmail user with a paid plan. But recently two issues are annoying me enough that I'm considering leaving:
1. The incessant "Using Gmail to run your business?" upsells. No, I'm unemployed and this is a personal account. Unlike the AI upsell, I can't seem to dismiss this permanently. It just snoozes it until the next time I open it.
2. The Search bar has become dangerously glitchy (at least in Firefox for Mac) if you type fast and have keyboard shortcuts on. It lets me type 1 or 2 characters before it starts treating every character as a shortcut, inadvertently deleting, muting, archiving emails. Search is what sets Gmail apart and now it is unusable. I reported this bug to Google months ago and my patience is running thin.
I've always found the Gmail search bar glitchy on Firefox but never on Chrome browsers.
I'll be typing something in the search box and suddenly I'll end up navigating somewhere with the keyboard focus having gotten removed from there.
Could use a more reliable source for this report.
I paid for and tried Google AI for Gmail and was appalled at how bad it is. The product team there is really not executing well. I've switched now to Shortwave. It works very well and having LLM+RAG queries for 20+ years of email archives has been very helpful to me.
Something I noticed recently is that titles of youtube videos are automatically translated.
I'm convinced that nobody in Google speaks more than one language, otherwise they would have never done that. It is impossible to turn off the unsolicited translations in search, and now youtube. I'm scared that soon they will force up audio translation as well.
It's not only the titles, but the descriptions as well. And viewers can't turn this behavior off, only the channel creator can and most will not.
All this AI is great. What I really want is to clean up my inbox by seeing where the mails are coming from. What I want is `select sender, count(*) from emails group by 1 order by 2 desc` or the equivalent. But I guess that doesn’t juice the stock price like AI does.
I can see how it is useful: for some people, reading and writing emails is a dreadful activity. But it is also an important facet of every human activity -- communication. So, it better be intentional and clear. I have turned off all smart features and let me handle how to say what I want to say.
Man I was in a Google News Initiative meeting and they started with a "fun poll" where they asked how you'd feel if you read an email that you knew was written by AI. Lots of people didn't mind knowing that was the case, and it was mainly for the reasons you put out writing emails is legit not fun for a lot of people.
Man I was so triggered though, like old-man-yelling-at-clouds level. Then I went back through gmail and exactly one of the last 50 most recent emails I've received was not an alert or a newsletter or an invite. IOW out of my last 50 emails only one of them would've fit the bill for an AI-written correspondence taking the place of a human-written one. All that butt-hurt and it doesn't even matter.
So long way of getting around to saying that your point about intentionality and clarity is clutch, and if people can get that with AI-written emails maybe I can cool my jets.
One of my tech clients emailed me about a Roku problem and one of the AI suggestions in Gmail was to tell the customer to try unplugging a Roku for a minute and plugging it back in. I pressed that button, proofread the suggested text, and hit send.
In truth it felt both amazing and made me uneasy, for AI was encroaching on my career of telling people to reboot their errant device.
For anyone looking for a cross-platform, local-first IMAP email client that "just works" and doesn't jam AI down your throat, check us out:
https://marcoapp.io
i do about 95% of my emailing in thunderbird on my computer, and 5% in the gmail app on my phone or on the specific email's website. all imap so they're basically synced already. i'm wondering if your marco app will be useful for my use case, and what the benefits would be. i think i would lose some benefits such as thunderbird extensions. also, to use proton mail on my phone i think i need the protonmail app. i use the proton bridge to use it in thunderbird. would the marco app work with proton on computer but not on phone?
Is there a way to suspend/disable Gmail account in a way, that no bad actors could overtake it, after Google disables/deletes(?) it after period of two years?
I've seen no indications of this supposed feature in my own experience or other sources. The article does not provide any evidence to support its claims.
I'll save you a click:
> Gmail doesn't just offer to write your emails for you, they actually do it, and it's up to you to delete the text it wrote.
> Hard to make a screen shot to demo without revealing personal info. That's how awful this thing is.
> It reeks of desperation.
This is the whole article. That's it. No screenshot. No details. Nothing. Just three random sentences.
200+ upvote on HN btw.
You have to wonder what Gmail reports about you. Does it detect suicidal thoughts? Child sexual abuse? Antifa activity?
A client emailed me for a meeting at Monday 2pm. The Gmail AI immediately marked Monday 2am on my calendar and it cannot even be deleted.
It's amazon deliveries for me. Gmail gets confused about multi-item orders with different delivery dates, and I cannot click away the banners displaying on top of the mailbox. Really poor UI.
OTOH, it's nice to see that there is some innovation happening - eventually (I am an optimist) they will weed out the lousy stuff and keep the useful.
Wow i really hope this is a mistake, that sounds like a severe bug
Well it's not a "bug". Language models make mistakes; it's a fundamental part of how they function.
This isn't LLMs. Gmail has been creating calendar events from emails for a long time. The bug would be if the user can't delete the event.
What makes you think it hasn't long been using a large language model for it though? Maybe not Large Language Model, the term seems to have come with ChatGPT et al., but Google's been doing machine learning for a long long time; the transformer architecture paper (kind of the groundwork for the current wave) was published by a Google team in 2017.
A lot of new Google features are branded 'AI', because it's so hyped it has broad consumer awareness, but a lot of Google features for a long time have used AI and just been brandless or at least 'AI'-less features.
It started about a month ago for me. I am subscribed to emails from an IT as a Gig platform called Field Nation. Thought I might pickup a bit of side work but never did.
Recently it started adding them to my calendar and there is no way to turn off this feature without also turning off useful features such as package out for delivery notifications.
Bizarre, I can always delete things from my calendar. It sounds like an implementation bug. Did you report it?
Since when did Google pay attention to implementation bugs?
Stop using Google products.
Can it can help answer my questions on behalf of my clients that use gmail?
...I'm confused, is there something I'm missing in this post? I only see 4 sentences, with almost no information.
Where I work, we get tons of feedback from customers, basically saying, "How do I disable this new AI feature? I never asked for this." Yet, the guidance from leadership is very clear: "More AI!" I suspect it's like this at Google (and everywhere else), too.
Do one thing and do it well, you assholes
Be dumb pipes
Google has already arguably ruined Youtube, which they own, over a period of many, many years
It doesn't matter if the community does all the work, the company will MITM themselves to make themselves look more important than they are
Yeah, you gotta watch out for this stuff. The other day I was using Gmail when it just gave me a million dollars. I gave it to charity under an assumed name.
Can’t screenshot without revealing the name but I think it’s actually super cool of Google to do this. Just distributed charity. With no benefit to themselves.
They did say that they did it based on my Google Photos showing that I was really good looking (can’t share because of privacy) so there’s some privacy stuff there but overall I think it’s good.
Did the user accidentally keep hitting Alt+H? Or are they part of some experiment? Gmail writes nothing for me unless I click the "help me write" link. I also don't know why it is so hard to record a screenshot of the behavior, you can write it to example@example.com, and the draft email UI can be clipped from where it shows your email address.
> Gmail writes nothing for me unless I click the "help me write" link.
For me it sometimes (I'm not sure why, it isn't even "intelligent" about which emails have options, automatic noreply@ emails sometimes have it) has buttons to pre-fill a reply with a message out of a few choices. It's not unprompted, but I could see someone accidentally clicking it instead of the reply button.
Yeah, I don't know what he's talking about.
I see obviously AI-powered tab-completion suggestions, and the Alt-H prompt suggestion when first writing an email, but I've never had it actually insert text unprompted.
And yet google's spam filters still let through so much AI slop spam. Any email from a new sender should be suspect, and if it has an awkwardly phrased opt-out line ("if you'd rather not hear from me again, just shoot me a "leave out"") then it should presumptively be marked as spam.
Spam classification has virtually zero to do with the contents of the message.
I have the gmail app set to only notify me for "important messages," and it pretty reliably doesn't trigger a notification for this junk. I wish I could have it use that system (whatever it's using) to actually flag messages as spam/mass emails.
If it's a seemingly personal message from an unknown sender, send it to spam. If it looks like a mass email (that someone might have signed up for), then put it in the Commercial tab or whatever it's called, so I can set up a filter for it or unsubscribe.
I think letting your email app raise notifications is a mistake, and I can see how letting it do that would amplify the annoyance of any misclassified messages.
Until about 9 months ago it was not an issue. I'm glad the gmail app has the "important only" feature, but just wish it was given more power so that when I'm at my computer they are properly categorized there as well.
I'm truly thankful I deleted my Gmail account years ago and my smartphone has no account and no g-apps.
Do we have to riot in the streets before governments regulate this shit?
Why regulate it? You don't have to use it (as you clearly moved elsewhere). Out of curiosity, where did you move to?
Please do yourself and the world a measure of good, and stop using GMail. Giving someone a GMail address is like saying "Yes, I like to be abused, I like to be violated and have no privacy". It is quite embarrassing - and people who are tech-savvy should be telling their colleagues and peers that it is embarrassing and inappropriate.
Yes, I realize that dropping the GMail mailbox does not mean you are free from GMail, because of all of the other people on it. But we must each make some effort to chip away at that thing. Twenty years ago, when Microsoft's hold on PCs was much stronger than it is today, online venues were inundated with derision and denouncement of them and their practices; but these days - it's as though Alphabet/Google control of all these aspects of so many people's lives is just neutral reality, ho-hum, nothing to waste time thinking about.
I’ve found the email thread summary pretty useful ymmv.
ha ha, hey google PMs, here's a useful AI feature:
> Is this e-mail marketing something?
I recently noticed that some of the emails in my gmail "promotions" inbox aren't emails at all, and are actually undeletable ads from Google.
If you're willing to use the web ui rather than the native app, then uBlock Origin will successfully remove those ads.
Otherwise, you could always just switch the "promotions" tab off - the contents will show up in your main inbox, but without the ads.
I moved to ProtonMail right before their CEO started mouthing off about his dipshit opinions. I still use them because it's cheaper to pay for a family plan and get a VPN, password manager, and storage along with the email. I used to pay for each of these services individually and was paying 2x what I pay now
> I used to pay for each of these services individually and was paying 2x what I pay now
What's the other factor though and is it worth the getting walled in?
Password manager in my opinion has to be its own thing, has nothing to do with the rest, same for VPN. I can see email, calendar and data being close together as being beneficial in every day flow. That's where Fastmail for instance draws the line I think.
If you untangle these, would the cost not be worth the benefit?
ProtonMail also does this. Someone actually told me they got an email from me and when I checked it said “this mail was AI generated”. I can’t screenshot it though. Personal reasons.
This is why I am building a better AI for my inbox.