News for hackers. Stuff that matters.

photoGrant

Re:AI behavior guardrails should be public on Thursday February 22, 2024 @12:42PM (#39470438)
I’ll be more subtle in my agenda to match the spirit of the site
Re:AI behavior guardrails should be public on Wednesday February 21, 2024 @08:40PM (#39462074)
:)
Re:AI behavior guardrails should be public on Wednesday February 21, 2024 @04:42PM (#39459979)
Then who in this black box private company is the Oracle of infinite wisdom and truth!? Who are you putting in charge? Can I get a vote?
Re:AI behavior guardrails should be public on Wednesday February 21, 2024 @04:21PM (#39459723)
This is written nonsense.
Re:AI behavior guardrails should be public on Wednesday February 21, 2024 @04:19PM (#39459697)
If it's a question of facts, why are you allowing blind assumptions to lead your opinion? Do you have sources and evidence for their agenda that matches your beliefs?
Re:AI behavior guardrails should be public on Wednesday February 21, 2024 @04:16PM (#39459665)
I've been part of the advertising and marketing world for a lot of these companies for a decade plus, I've helped them sell bullshit. I've also been at the start of the AI journey, I've downloaded and checked local models of all promises and variances.

To say they're better than the compute that OpenAI or Google are throwing at the problem is just plain wrong.

I left the ad industry the moment I realised my skills and talents are better used informing people than lying to them.

This thread is not at all comparing the ethical issues of AI with local anything. You're conflating your solution with another problem.

Re:AI behavior guardrails should be public on Wednesday February 21, 2024 @04:12PM (#39459626)
I never said each of these exist in a vacuum. It is the collation of all that is the danger. This isn't democratic. This is companies now toying with governmental ideologies.
Re:AI behavior guardrails should be public on Wednesday February 21, 2024 @04:10PM (#39459602)
No, ignoring laws and stealing data to increase your Castle's MOAT is the win. Compute isn't an open source solvable problem. I can't DirtyPCB's an A100

Making the argument open source is the answer is an agenda of making your competition spin wheels.

Re:AI behavior guardrails should be public on Wednesday February 21, 2024 @03:57PM (#39459443)
No, compute is keys to the kingdom. The rest are assets and ammunition. You out-compute your enemy, you out-compute your competition. That's the race. The data is part of the problem, not the root.

These companies are silo'ing the worlds resources. GPU, Finance, Information. Those combined are the weapon. You make your competition starve in the dust.

These companies are pure evil pushing an agenda of pure evil. OpenAI is closed. Google is Google. We're like, ok, there you go! Take it all. No accountability, no transparency, we trust you.

Re:AI behavior guardrails should be public on Wednesday February 21, 2024 @03:39PM (#39459197)
Absolutely it won't. We've armed the issue with a supersonic jet engine and we're assuming if we build a slingshot out of pop sticks we'll somehow catch up and knock it off course.
Re:AI behavior guardrails should be public on Wednesday February 21, 2024 @03:22PM (#39458960)
Remind yourself we're discussing censorship, misinformation, inability to define or source truth and we're concerned on Day 1 about the results of image gen being controlled by a for profit single entity with incentives that focus solely on business and not humanity...

Where do we go from here? Things will magically get better on their own? Businesses will align with humanity and morals, not their investors?

This is the tip of the iceberg of concerns and it's ignored as a bug in the code not a problem with trusting private companies with defining truth.

Re:Ask HN: Who is excited for OpenAI's Sora? on Wednesday February 21, 2024 @03:15PM (#39458898)
AI Profiteers are excited. Everyone else is annoyed.
Re:Reddit sells training data to unnamed AI company ahead of IPO on Wednesday February 21, 2024 @02:56PM (#39458668)
No one accepts this is ok. Personal agenda might argue people should've known better. Something something Spanish inquisition.

The same crowd here that argue EULA's are a joke and need refactoring to protect the consumer are forgetting they ever argued this fact a day in their life.

Re:AI behavior guardrails should be public on Wednesday February 21, 2024 @02:53PM (#39458630)
Oh?

https://news.ycombinator.com/item?id=39449887 - nerf'd https://news.ycombinator.com/item?id=39406709

these discussion are buried by fingers in ears thinking.

Re:AI behavior guardrails should be public on Wednesday February 21, 2024 @02:52PM (#39458618)
I've had 'AI Attorneys' on Twitter unable to even debate the most basic of arguments. It is definitely a self fulfilling death spiral and no one wants to check reality.
Re:Tinder Owner Signs ChatGPT Deal. Enjoy the AI Dating Tidal Wave on Wednesday February 21, 2024 @02:45PM (#39458550)
I have more faith in society than I do the vultures eating away at it, at least.
Re:75K loyal Redditors can snag shares before Reddit goes public on Wednesday February 21, 2024 @02:45PM (#39458541)
How altruistic! This definitely has nothing to do with keeping the biggest court case arguments at bay.
Re:Tinder Owner Signs ChatGPT Deal. Enjoy the AI Dating Tidal Wave on Wednesday February 21, 2024 @01:59PM (#39457962)
This arrogant vacuum is going to get checked by society.
Re:What Was Sora Trained On? Creatives Demand Answers on Wednesday February 21, 2024 @01:17PM (#39457386)
It's cognitive dissonance to such an arrogant and ignorant degree it's turning out to be the best sunlight we've ever needed on the situation. No one is arguing a balanced or logical alternative. Every argument ignores the entire landscape of issues and reduces it to 'what's yours is mine but what's mine is not yours'..... They're arguing to keep the toys in their playground. Even if it's not their toy. It's 'in the world' so we're free to take it.

Ok! I appreciate this insight. At this point I'll be 'scraping' everything from tech and fight this in court with their own words.

Oh, I'm sorry, did I argue my omnipotence? Or the fallacy of marketing? You shifted the goalposts pretty far.

Our existence in the 'now' means we're reactive to the now and the universe of events influence us as much if not more than our concept of free will. I don't think to make my skin react to cold air. If these models can't ingest, quantify or even understand/intuit current stimuli, they're a dead end solution and the problems are bigger, harder and still being ignored during the cash grab era.

I haven't seen or heard a single word of progress on that front. Any sources?

We agree. But remember, approximations here are nothing but lies masked as convenience.
Re:OpenAI talk on Weak-to-Strong Generalization [video] on Wednesday February 21, 2024 @04:44AM (#39451858)
"For the purposes of this talk the details don't matter"

So if you want to learn the context of what 'performance' means there's no answer.

"You shouldn't trust any of these numbers too much" but infers a trend.

Dogfooding Era

Re:Corporate Responsibility in the Age of AI on Wednesday February 21, 2024 @04:25AM (#39451751)
I'll save you reading;

> With that in mind, we need to ask what companies must do to use AI responsibly.

> What are a corporation’s responsibilities in the age of AI—to its employees, its customers, and its shareholders?

Responsibility and ethics stop at the incentives of the business and don't stretch one iota further.

Ethics is a marketing word

Re:Pentagon explores military uses of emerging AI technologies on Wednesday February 21, 2024 @04:11AM (#39451673)
Where's my opt out.
Re:What Was Sora Trained On? Creatives Demand Answers on Wednesday February 21, 2024 @03:40AM (#39451458)
Currently the weapons against bullshit is trying to talk to it. Engage and you'll find only 5% of the time there is depth, or earnest.

The rest is just airing itself like dirty laundry by the second response and ultimately just serves in contributing to a humans ability to quickly adapt to new environments. Which means bullshit just smells more nuanced today.

To model physics and not just a playground pretend version of it, you need to model the complete understanding of physics within the entire universe. We'd need answers to questions we're UNKNOWN-TIME away from understanding. This means modelling everything from the visual response of a human eye to the known-truths of the universe and beyond.

We can't even trust our own nose to smell bullshit but we're supposed to have a model for it.

Please think critically folks!

Re:What Was Sora Trained On? Creatives Demand Answers on Wednesday February 21, 2024 @03:20AM (#39451340)
Hindsight being the premise for confidence in this context is precisely why the hubris of factual speaking on the what-if's of the future can be pretty easily ignored as self infatuated authority. The rest of your arguments are still conflating the problem. One enhances the voice of an individual the other extinguishes it.

Regardless your semantics on your phonograph argument the reality remains. The artist remained as part of the contract of societys growth. Once again, this isn't a blank vinyl or a canvas or an empty parking lot where freedom of expression takes priority. The priority is the financial incentives and retaining control.

For you to argue there's any creativity that would come out of that is like arguing I'm being creative when I put coins into a vending machine and receive cigarettes. I'm as creative as the conditions the puppets of profit have allowed me to be. Money doesn't own this world despite its deep and confused hope. It steals from it.

> They spoke in terms exactly like you are now, and those ideas were wrong then, and they are almost certainly wrong now.

No, they were right then and they're right now. Otherwise based on your argument anything beyond a phonograph would be unnecessary. So why do people seek anything beyond their MP3? Why do they go to live shows? Why do people still visit the Mona Lisa even though they can see it at home for free with a google? Please, as much as you refuse to debate philosophy and art, those are the exact things being debated. So either listen to those who've spent their lives in their domain (it's why I don't falsify my Python accolades) or pretend you know better. But don't conflate. Do read Walter Benjamin's essay from 1935 on Works of Art in the Age of Mechanical Reproduction. I know, you won't.

Next you're going to tell me AI hip-hop has the same value, incentive and understanding as the people from Cabrini Greens trying to escape poverty and violence by creating something beautiful and relatable out of the human condition. Then you'll tell me it's not your fault they haven't 'out created' themselves to keep up with the theft of their very self. If the vultures could at the very least just fuck off until the person isn't alive enough to bare witness to it, you could at least feign some IP argument. It's otherwise just hot air hoping.

*ALL* current public proof of concepts for AI are a dead end. They are strictly past information ingest machines. Until 'Generative' solution thinking finally stops chasing the tail of history and believing that what it has ingested from then to now is just as simple as adding more 9's... and the solution will manifest -- is ignoring that without the ability to also quantify stimuli in every definition - of the 'present' - it's nothing but a well dressed toy. We can't even define 'present' and yet we think a farm of GPU's will magically solve singularity. If anything, this is infinitely short sighted.
Re:What Was Sora Trained On? Creatives Demand Answers on Wednesday February 21, 2024 @02:09AM (#39450954)
Yes. Master.
Re:What Was Sora Trained On? Creatives Demand Answers on Wednesday February 21, 2024 @02:01AM (#39450903)
One is an expression of all through self, another is a skill to a solution. One is given value by society's response, the other by a companies incentives.

When you take an artists desire to speak to the world through self expression, and believe it's equivalent to building software you're forgetting that one of them is already extremely well compensated, the other is not aligned with those incentives.

But to be extremely well compensated, and then try to rob artists of their livelihoods for profit and convenience -- yes, folks are going to think this argument isn't quite as balanced as you're painting it.

Older »