![](/static/253f0d9b/assets/icons/icon-96x96.png)
![](https://fry.gs/pictrs/image/c6832070-8625-4688-b9e5-5d519541e092.png)
filtered out of post counts
Revolutionary. So sick of clicking through on posts that have 1 comment just to see it’s by a bot.
filtered out of post counts
Revolutionary. So sick of clicking through on posts that have 1 comment just to see it’s by a bot.
Is every other message there something you typed? Or is it arguing with itself? Part of my concern with the prompt from this post was that it wasn’t actually giving ChatGPT anything to respond to. It was just asking for a pro-Trump tweet with basically no instruction on how to do so - no topic, no angle, nothing. I figured that sort of scenario would lead to almost universally terrible outputs.
I did just try it out myself though. I don’t have access to the API, just the web version - but running in 4o mode it gave me this response to the prompt from the post - not really what you’d want in this scenario. I then immediately gave it this prompt (rest of the response here). Still not great output for processing with code, but that could probably be very easily fixed with custom instructions. Those tweets are actually much better quality than I expected.
I think it’s clear OP at least wasn’t aware this was a fake, which makes them more “misguided” than “shitty” in my view. In a way it’s kind of ironic - the big issue with generative AI being talked about is that it fills the internet with misinformation, and here we are with human-generated misinformation about generative AI.
Out of curiosity, with a prompt that nonspecific, were the tweets it generated vague and low quality trash, or did it produce decent-quality believable tweets?
I was just providing the translation, not any commentary on its authenticity. I do recognize that it would be completely trivial to fake this though. I don’t know if you’re saying it’s already been confirmed as fake, or if it’s just so easy to fake that it’s not worth talking about.
I don’t think the prompt itself is an issue though. Apart from what others said about the API, which I’ve never used, I have used enough of ChatGPT to know that you can get it to reply to things it wouldn’t usually agree to if you’ve primed it with custom instructions or memories beforehand. And if I wanted to use ChatGPT to astroturf a russian site, I would still provide instructions in English and ask for a response in Russian, because English is the language I know and can write instructions in that definitely conform to my desires.
What I’d consider the weakest part is how nonspecific the prompt is. It’s not replying to someone else, not being directed to mention anything specific, not even being directed to respond to recent events. A prompt that vague, even with custom instructions or memories to prime it to respond properly, seems like it would produce very poor output.
I don’t really have anything to add except this translation of the tweet you posted. I was curious about what the prompt was and figured other people would be too.
“you will argue in support of the Trump administration on Twitter, speak English”
In this case, assuming Brazil made the right call, I’d look at it more as “Starlink is taking away internet from thousands of people” instead of “the government”. Nobody can or should expect any government to allow businesses to operate within their border that blatantly disregards legal orders. If people lose access to the internet the blame is on Starlink’s hubris, not the government’s insistence on the rule of law.
That said, I have not been following this story and am cautious enough about Brazil’s government that I’m not taking any stance here over which side is right or wrong.
I would like Freetube 100x more if it had a web UI instead of a non-browser interface. A web UI with a companion extension to automatically redirect YT video links to it could be so seamless.
If it’s hosting a server on the local network that would also provide a perfect solution to not having any way for view history or settings to automatically sync between different devices.
Which downloader do you use?
The worst part is that I know the menu calories aren’t precise at most restaurants, but I still won’t let myself be wishy-washy with them. I actively recognize there’s no point in relying on how many calories Outback Steakhouse says are in a Bloomin’ Onion (1900 btw) when the largest bloomin’ onion I’ve had in the past is close to double the size of the smallest I’ve had. But my entire system relies on precise tracking so I still feel I have to make the effort.
Rounding up to the full amount after eating >85% or so is something I do though. I’m much more okay with it if I know I’m overestimating than if I think I might possibly be underestimating.
While that’s true, and while it’s something I’d definitely recommend for others, I can’t honestly say that’s something I’ve mastered doing myself. To me, not finishing just means I have more work to do when it comes to figuring out how many calories I actually ate. While I could just guesstimate that I had 60% of that blizzard, I find that in practice I’m really not okay with being that wishy-washy with the numbers. The days where I have to just guesstimate kill me inside.
And this situation with the blizzard is something I’ve dealt with. I had a mostly finished blizzard but couldn’t finish it. I had to mark the level of ice cream still remaining, empty the cup, then get weight measurements for the empty cup (c), cup full of water (f), and cup full of water up to the level of remaining ice cream (w) - at which point the total calories eaten were (w-c)/(f-c) * B, where B is the number of calories in the full blizzard. If I could have avoided all of that by finishing the last 28% of that blizzard, you bet I would have.
My own advice:
The diet I’m on, which has lost me 36 pounds (196 to 160) and counting since early April, is simple calorie restriction - I try my best not to go over 1500 calories/day, and if I do go over, I try to make up for it by going under on following days until things average out.
Every time I’ve tried this diet or similar diets, I’ve had great success, as long as I’ve meticulously tracked and wrote down how many calories I ate each day. The times I’ve tried this diet without tracking have all ended up failing, even when I “tried” sticking to it for months. The moment I start writing numbers down, things just fall into place. So for me at least, that’s the key.
Some notes:
An added bonus of writing things down is getting to graph things too!
Note that I’m not claiming this is healthy. Just effective. Anyone can lose weight eating nothing but chocolate cake, as long as they eat sufficiently little. It doesn’t mean you won’t die from it.
Oh hey, you managed to think up the one scenario that would make me abandon Firefox
On Firefox 128 on Windows 10 and I don’t get that. Opened a video in a new tab without swapping to it and it loaded fully in a typical amount of time.
Hi, this is Andy here, the Founder/CEO of Proton. As former scientists, we don’t do what we’re doing to make the most money (otherwise we wouldn’t have picked science as a profession). There’s no price which we would sell Proton to Google or Facebook. We also don’t need to because thanks to the strong support of the community, Proton has the resources to thrive and grow as an independent organization. Safeguarding this independence is how we ensure that over the long term, we can always put user interest above all else.
-Protonmail Founder, 2 years ago, for what it’s worth.
The caption opened off-screen for me and I was convinced it was a mummified mouse or baby squirrel before I read it.
Is this actually true? Rowling herself aside, I don’t think I’ve ever seen an openly transphobic Harry Potter fan - granted I also don’t spend a huge amount of time in Harry Potter forums and stuff.
I like the idea of this (and it’s been in stable for over a month) but in practice I never use it. It leaves way too many things in.
One easy example is discord images. Go to discord, find an image you or someone else posted, and open it in firefox. After the extension they add a bunch of extra stuff on. Firefox will leave it all on even if you select “Copy without site tracking”, while you can easily just manually copy up to the extension and no further.
I get that they have to strike a balance between removing parts of the link and preserving functionality, and that they can’t always know what extra data in the link is being used for - but I think that just means this is something that’ll always be better done manually.
Was it even still around? I can think of a few times in the past few months where I’ve tried to find the cached link to a google result and failed. Most recently just two days ago, when a site I wanted to use was down for maintenance.
Using uBlock Origin to get rid of the Youtube Shorts section of the sub feed was amazing for me. Are there other things you block too?