This is a most excellent place for technology news and articles.
'Make this chat discoverable.' Beneath that, in smaller, lighter text, was a caveat explaining that the chat could then appear in search engine results."
UX designer here. People don’t read the little gray supporting text. “Search Engine” should’ve been in the headline.
I've always been under the impression that the little grey supporting text being little and grey is because the designer didn't want it read but was required to put it somewhere. A dark pattern, if you will. Is it actually not intended that way?
When it’s used correctly, it should be adding a little extra color or context that’s not critical for most users, but will be helpful to a certain segment.
Or it’s bullshit that you -know- the user doesn’t care about, but it’s needed to make some person or department happy.
Or it’s a dark pattern.
the ignorance required to lean into 'AI' in such a way all but ensures this
it's pretty telling that I hear AI talked about and seemingly used most by conservative types, I imagine because it's being pushed by influencers in that realm and the same people bankrolling 'AI' everywhere. people that are already comfortable with blind faith.. makes a bit of sense
then you still have people who are used to challenging and questioning things still upholding skepticism and not trusting ai because it all reeks of shit
AI for next pope!
That and conservatives abhor thinking. They need someone or something to defer to. And LLMs can give them all sorts of moderately intelligent bullshit they've come to expect . From their leaders and politicians. But better and faster. With possibly even less accuracy.
Apparently these people ticked a box saying "allow this chat to be indexed by search engines" and were surprised when their chats were indexed by search engines?
Are these the same users that think the sycophant machine really loves them?
Well I'm shocked more people don't ignore the flashing lights at train stations and just drive into the tracks right in front the trains frankly.
They kind of do.
unrelated but kinda related - I’m a firefighter. Often seen folks driving toward the scene lights like moths. Can be sketchy sometimes. I suspect it’s the collimated leds just piercing into the brains of the already distracted drivers.
Target fixation. If you're looking at something, you drift towards it.
Seems like a skill issue to be honest. Bunch of boomers checking boxes they dont understand, since thats entirely optional thing to create. Those public links arent created by themselves.
Says in the article the users clicked "share chat" then shared the links with others on services like What's App.
Sounds like What's App should get some flack for this too.
I don't see why, least my understanding, you hit share chat, it creates a public link... google's robots discover everything public and index it. Seems to me like the same problem would happen if you generated a link to share on any platform, and burned it and never sent information to any platform. Unless googles indexing all whatsapp messages, but that would be a much bigger story.
Anyway point is blame IMO falls on either chatgpt for not properly configuring a robots.txt, or google for not following it.
How does it discover the link though? Is it crawling your whatsapp chats, or just trying every possible chatgpt share link?
Y'all need to read the 5 minute article. It was a short lived feature where they had a checkbox that said: "index my chat into search engines". Which is honestly dumb as shit if you ask me.
sounds to me just like googles bots are finding them... could also just be maybe chrome or similar taking note when it finds a page and dropping a dime to google.
I don't know what the deal is, just that the article specifically names What's App.
When you share something it's not private anymore! More news at 23:00!
Its one thing to not be private. Its an entirely different thing for that thing to be crawled, indexed and published on the world's biggest catalogue
When it's chats with LLMs trained on this very type of data, it's mostly the user's fault. Of course, executives of LLM companies should still rot in prison.
It’s totally avoidable if you don’t use it, but I think the onus is mostly on the companies for advertising these chat bots as like, a friendly personal assistant when that’s absolutely not what they are. Like all “AI” shit, it runs mostly on consumer deception.
Yep, and when you click a button that liteally says "make this discoverable on search engines" which is off by defualt, its the later.
Not on the internet it's not.
Oh no. Anyway.
My only thought is "no shit"
lol this was on purpose
how are they shocked, when they also get blog posts, and other posts being summarized on the AI search?
Omg! Now everyone will know about my erectile dysfunction!
FAFO.
Shocked, shocked I tell ya.
Insert Casablanca.gif
I mean, we knew this…
wait, even if I only used duck.ai?
No
it claims to be private, but also couldn't answer me when I asked how I could verify that claim
