5 ms·
> It’s getting hard to read HN comments about anything AI related with all of these comments substituting their own assumptions and assuming the worst case scen
by wokwokwok 3y ago
> It’s getting hard to read HN comments about anything AI related with all of these comments substituting their own assumptions and assuming the worst case scenario without any effort to read the article
Maybe there’s a reason for this beyond bias?
I mean, HN is a reasonable forum. It’s got a pretty good set of moderation standards. The folk here aren’t particularly prone to take sides blindly without thought or reason.
If you seeing this (and I’m not, particularly, but let’s pretend for a moment you are); maybe AI has an image problem.
Maybe? I mean, what are you looking for? What more from a community like this would you like to see?
Here’s my $0.02:
I use jetbrains products all the time.
I read the article.
I am not anti-AI.
I’m still uncomfortable with it; because today when I was using rider, I got a feedback pop up saying “how would you rate this AI completion?” from a normal completion.
I don’t have assistant enabled. I’ve never used it. I haven’t signed up for it. I’m not using an EAP.
It was probably just a screw up of some kind.
…but, when you deeply integrate this stuff into the platform, that’s what you get; you cant tell when it’s active or not.
It should be a plug-in.
I don’t like it when, for whatever technical reason, I’m forced to use an AI product, when I didn’t opt into using it, and I’m not satisfied, based on my personal experience that they’re doing this right.
Maybe anti AI sentiment you see is more considered, in some cases, than you believe.
- MiddleEndian 3y ago>I’m still uncomfortable with it; because today when I was using rider, I got a feedback pop up saying “how would you rate this AI completion?” from a normal completion. Quite honestly if I got a popup like that from any text editor I would never use it again.
- johnnyanmac 3y ago>If you seeing this (and I’m not, particularly, but let’s pretend for a moment you are); maybe AI has an image problem. Clearly. Look at the players and where the money is, and look at how those players were doing with other tech last decade. No surprise that their goodwill from the get go is not only drained but in debt. Even before going into the actual ethical and logistical quandaries of AI, I was already skewed to be skeptical. So it is a shame when you see similar mannerisms dropping from companies you did trust. again, even without considering the idea of AI: the company provided a required update that added a new plugin in by default, which leads me to another subscription. This is how the nickel-and-diming starts.
- PH95VuimJjqBqy 3y ago> I mean, HN is a reasonable forum. not how I would describe HN.
- Obscurity4340 3y agoHow would you describe it?
- PH95VuimJjqBqy 3y agoIt's a safe space for people with money to express opinions that would otherwise get them excoriated. In many ways it's the online version of the upper class insisting that the silverware must be arranged in a certain way and used only on specific types of food with specific ways of holding them. It's mostly fake and one of the ways you can see this is the sheer amount of passive aggressive behavior witnessed.
- iamjackg 3y agoThey have added a small local-only model that runs directly on your machine to try and do smarter autocompletion. The model only runs if your machine is considered powerful enough to provide a sensibly fast experience. Nothing ever leaves your computer. I was equally confused, then I found all these details somewhere in their documentation.