Just leaving a feedback comment about the replies on this thread: I am surprised that people in this community disliked such a breakthrough in AI technology.
A lot of people have a hate boner for AI. They’re not interested in AI news, they just see AI and downvote.
Yeah, but that’s okay. What’s strange is the negative feedback in a community that specifically covers topics about AI.
Downvotes aren’t supposed to be used just because you don’t like, or personally disagree with something, but regardless, when a post makes the front page you’re going to be getting comments from people not subscribed to the community.
See the “fuck cars” community for example.
There is no “supposed to be used” for. They are meaningless internet points that we give to comments and posts for our own emotional enjoyment. Treating it as more than that is always going to cause problems.
It’s true that there are no official guidelines on the use of upvotes and downvotes, but if they are used to bury comments and posts just because they don’t align with your personal opinion, Lemmy is at risk of becoming a circle jerk and an echo chamber and I think that would be a great shame.
Above all else we should be encouraging participation, I don’t want to think that there are potentially great posts and comments out there but that people aren’t posting them because they’re worried about downvotes. The freedom of expression means that on occasion you might have to read things that make you uncomfortable and that’s OK.
Lemmy is at risk of becoming a circle jerk and an echo chamber
I’m not sure anyone aspires for it to be anything more than that.
people aren’t posting them because they’re worried about downvotes.
Such people are looking for a different medium. This is ersatz Reddit. We downvote. It’s fundamental to the culture and the format. It doesn’t have to appeal to everyone. There’s the whole rest of the Fediverse, where AFAIK you will not find a single downvote.
Maybe I’m only speaking for myself, but as far as I’m aware my opinion isn’t unique and I’m absolutely not here for a circle jerk or an echo chamber, they were my least favourite things about Reddit.
I’m aware of how people use downvotes in reality as opposed to how they would use them ideally in my opinion, I just think it would be a bit of a shame to end up where we left off if we have the opportunity to improve, especially if it means that new or controversial ideas are shared more freely and debate is a lot more open.
Put that back in your pants, this is no breakthrough
How did it get concrete requirements out of a customer?
I’d love to hear from anyone who has actually tried this (and fully expect my boss to put a team on it once he learns about it) as it sounds too good to be true.
The person in this video already has access and demonstrates how this new technology works. He is not just a random youtuber. He actively delves into AI news and even studies research papers.
So what’s the problem and how did it get solved? I can’t be arsed to watch a video that’s probably just ai shilling
Spoiler, it’s not just ai shilling.
OK I watched this so that everyone else doesn’t have to.
It’s about AWS Bedrock Guardrails.
First he says that AWS IAM which is just role based access control for their cloud is something insurmountable for a single person to accomplish (which is bullshit, I have implemented this myself in some projects).
Then he goes on to shill this new product where they use a LLM to take any policy document and make it into something logical (and pointing out any logical inconsistencies while also allowing you to correct any assumptions it made). This is actually good because a lot of policy documents are vague and conflict with themselves.
The product seems useful but it’s far from ground breaking and his way of talking is sensationalist.
