Live data from Hacker News

Flux 3

bfl.ai

81–90 of 151 posts

Re: Flux 3

#81

Earlier quoted context omitted.

As I understand it HN is a community for hackers to discuss interesting and curious topics. Pessimistic opinions on the labor market, views about society and politics that border on the dystopian, as well as exaggerated concerns about datacenter environmental impacts, all seem like a poor fit for this website.

It would be a very uninformed world view and naive to say that the people who build and implement cutting-edge technology shouldn't be thinking about its societal implications and dangers. Or even considering more basic things like making sure it doesn't hurt anyone. Just because you can do a thing doesn't mean you always should do a thing. Anyone ever taken a CS course that includes extensive discussion of the Thera…

My issue with comments around the impact of AI is that often they are just not interesting.

What substance is there beyond "I don't like it, and I think it is a negative for society", typically delivered with what I perceive as a self-righteous attitude?

Re: Flux 3

#82

Earlier quoted context omitted.

I'm a person who is making extensive use of current-gen "smart" LLM for coding tasks and building automation tools, doing some of the drudge work of gluing disparate open source things together into new projects. I'm fairly optimistic about that. Particularly when you have a good harness setup and you know enough about the subject matter to understand when a model has gone into some dead-end of reasoning or has built…

Only the enlightened tech elite should have access to such sophisticated technology - we can’t expect the uneducated masses to not misuse them /s

Yeah, really, masses are fine. I mean gross and boring mostly, but fine. Problem is when entrepreneurs get their hands on such tech, because they invariably put it to use for scamming and cheating people (read: marketing). It's the business people who are flooding social media with AI slop, for example.

Re: Flux 3

#83

Earlier quoted context omitted.

It would be a very uninformed world view and naive to say that the people who build and implement cutting-edge technology shouldn't be thinking about its societal implications and dangers. Or even considering more basic things like making sure it doesn't hurt anyone. Just because you can do a thing doesn't mean you always should do a thing. Anyone ever taken a CS course that includes extensive discussion of the Thera…

My issue with comments around the impact of AI is that often they are just not interesting. What substance is there beyond "I don't like it, and I think it is a negative for society", typically delivered with what I perceive as a self-righteous attitude?

My criticism is specifically with the use of text-to-image and text-to-video generators that will create fully artificial video and image content. I've seen the uses that these tools are being put to in 2025/2026 for electoral manipulation, social issue manipulation and it leaves me with a very distasteful impression.

You will not see me leaving comments such as this, for example, on the release of a "strong in coding tasks" tool like GLM5.2.

I actively go out of my way to avoid patronizing businesses that advertise with AI slop generated images now, AI slop restaurant menus, and so forth.

Whether you want to interpret a desire for authenticity as some sort of self-righteous attitude is up to you. There's a lot of people that share my opinion, and a lot of them that don't. There is also clearly a lot of money behind pushing AI slop images everywhere. Facebook and the various 'pages' and 'groups' that are near 90% AI slop content are a fine example of that. I'm sure a great many advertising impressions and click-throughs have been served, much revenue has been earned. Great success.

Re: Flux 3

#84
post #26

Earlier quoted context omitted.

none of this is exciting compared to holodeck in star trek

> in star trek Heh. On the other hand, we are seeing so much negativity around "AI", while we have access to something as close as possible to the "universal communicator" from trek. There are now several CotS things you can buy that basically serve this purpose. You can walk into a mall, buy one, and travel to the most remote place on Earth, speak in your own language, and almost anyone else can speak in their own l…

Yup. People still don't believe when I say that we've cracked universal translation with LLMs. That's despite the proof being in their hands; anyone with any SOTA multimodal LLM app (read: Claude, ChatGPT, or Gemini) with "advanced voice" / "live mode" can test it on the spot: just launch it and tell it will hear multiple languages and its task is to translate what it hears in one language to the other, and vice versa.

Re: Flux 3

#85
I hope flux will include a 3D generation model. right now the open-weights version of 3D is failing behind closed source by a big margin. Hopefully the improved spatial ability helps with robotics too

Re: Flux 3

#86
post #28

Earlier quoted context omitted.

Open-weight promise seems nice, but 1) I've seen people commenting that these hopes amounted to nothing for some previous releases (no idea which promises were made though); and 2) if the backbone is released as Dev, what will be missing? I can't easily tell from the post.

I run some of the 2.3 models locally so I'm not sure where you saw that they didn't follow their words?

It is not that they don't release open weights, but some users report that they are significantly inferior to the closed versions.

Re: Flux 3

#87
post #32

i don't get why they are investing money on image/video gen. All generations i have see looked blurry, lacking in fine details and missing the artistic touch (lacks meaning? lifeless?)

[dead]

Re: Flux 3

#88

Earlier quoted context omitted.

My issue with comments around the impact of AI is that often they are just not interesting. What substance is there beyond "I don't like it, and I think it is a negative for society", typically delivered with what I perceive as a self-righteous attitude?

My criticism is specifically with the use of text-to-image and text-to-video generators that will create fully artificial video and image content. I've seen the uses that these tools are being put to in 2025/2026 for electoral manipulation, social issue manipulation and it leaves me with a very distasteful impression. You will not see me leaving comments such as this, for example, on the release of a "strong in codin…

> electoral manipulation

That must be the poster child of the most boring and meaningless accusation thrown around new technologies (after the data center water scare, which was just plain bogus). We've heard this non stop since Cambridge Analytica nothingburger. It's all noise distracting from the one meaningful aspect of it: marketing to people is electoral manipulation, and since that's unquestionably allowed, the horses have left the barn long time ago, and futzing over the AI barn door control makes no sense at this point.

If anything, the problem that this is effective in the first place is the one to address - this translates to people still believing anything politicians say, despite decades of continued proof all the campaign promises are just plain bullshit, and stated beliefs are situational and not principled.

Re: Flux 3

#89
post #47

AI slop trained on copyrighted content.

"Horseless slop trained on centuries of coachmakers' and blacksmiths' work. It'll never replace a real horse." - Hacker News commenter in 1889 criticizing the automobile

This is close to the stupidest thing I've ever read. Congrats

Re: Flux 3

#90
Awesome -- glad they're going to release the open weight version! I've been waiting for an excuse to re jump into AI OS image gen!
Post reply on HN