mrdependable

Don't you need to look at the code you want to change in order to direct the AI in what you want to do, or are you simply telling it to make a feature and then looking at the changes it makes?

I think of LinkedIn like walking into a Tony Robbins event. At least, what I imagine they are like from those old infomercials.

A sad state of affairs when the law is what people look to in order to decide between right and wrong.

If I'm not mistaken, isn't this the same person behind Lambda School? I saw the same ad like a year or two ago.

Hobbling the model may be smart tactically for them, but feels like it sets a really dangerous precedent.

I suppose if Spaghetti Inc and Olive Garden were in the business of creating and licensing recipes, they would both be in trouble. But they are selling labor and ingredients, which is what people are paying for. Also, pizza and spaghetti have been around a long time and patents don't last forever, so I don't see how this applies.

Maybe a better example would be to explain how drug companies are able to cover the cost of research and development, or how Studio Ghibli is able to invest in animating a movie that anyone can sell. Should everyone provide the full suite of services a business needs in order to earn a profit? That seems like an inefficient way to organize the economy. The mechanical engineer also has to be a marketer or else he doesn't deserve to be paid for an invention? How does that make sense?

Sure, different inventions work differently. What does that matter?

You cut the sentence that answers the question.

An AI training is not equivalent to a human learning, despite how we like to anthropomorphize the process, and ideas are not protected by copyright in the first place.

Your point was that people that create something valuable are not entitled to stop others from capturing the value of their creation, ostensibly because they “don’t want to do the hard work.” My point is that even if someone did want to do the work, the benefit would go to the people with the money and infrastructure to act on it. In your vision, IBM would just take Windows and slap their name on it. Where would that have left Microsoft?

Computers and cars don’t rely on the intellectual output of others to function. AI still needs people doing the work that it is displacing, but it strips the economic incentive to do it.

You say means of production is an antiquated idea, but the world you are using as an example of the idea being unnecessary is a world where copyright exists. The world would look very different otherwise, and you can actually get a sneak peek of what would happen by looking at AI companies. These are not people using ingenuity to capture value that no else bothered to. They are owners of massive capital training for free on everyone's work to not only capture their value in ways the little guy can't, but to replace them outright.

What about something like carbon credits? Every citizen gets awarded a yearly AI usage credit. If a company wants to use an LLM, they have to buy the usage from the public market. People can use their own credits freely.

It's entitled for others to believe they have any say in what happens with the work of others. The world without these protections would be worse off by far. What I gather from what you are saying is that if I write a song, or a book, anyone else should be able to take what I've done and make their own money off it. By that logic, a publisher wouldn't need to compensate writers. Record executives wouldn't need to compensate musicians. Whoever holds the means for extracting value are at a huge permanent advantage.

We haven't gotten to where we are in the world today by giving the wealthy huge permanent advantages. Look at the explosion in innovation that has happened since public education has become widespread. It used to be only the wealthy that could afford to be educated. Part of what makes capitalism work at all is by not allowing the means to capture value to be monopolized.

I wonder if this could be explained in a similar way to Hollywood movies. If the movies are designed to please the largest group of people, there is a greater chance people will choose to see it than another movie. The human law professors come with their own personalities, beliefs, and opinions that come through in their writing. An LLM has been trained to please the largest swathe of the population. That doesn't mean the answer is better; just like Captain America isn't necessarily better than American Beauty.

How do you envision that playing out? It would basically be like everyone that didn’t still have a job living off minimum wage. Would no one be allowed to work also?

You may not need to have a job to be happy, it varies person to person. However, the idea that the billionaires will save us and our leverage is not needed is ridiculous. It is much more likely we would see poverty like is seen in much of the rest of the world.

I think a lot of this would be solved if the government would actually enforce anti monopoly laws. The penalties against Google were such a damn joke it makes me sick.

Companies aren’t going to hold off on trying to automate tasks AI performs poorly at. They are going to change the task so AI can do it, putting the burden of making up for the deficit on everyone else. The only reason this is able to happen now is because all the competition has been crushed or absorbed.

You are ignoring the part where human labor is the leverage required for democracy to work.

I am not talking about how much data is required to make intelligence. I am talking about how it uses the data it already has. It can tell you about every scam in the book, research about the scams, how to spot scams, who does the scamming, etc. Everything under the sun about scams. However, without the “skill” included in a prompt it will fall for scams.

With the amount of data these models have, they should be much more capable if there was an actual intelligence behind it. If you saw someone running into a wall continuously until you showed them how to use a door, even though they have seen people use doors a million times, what would you call that?

The fact that Anthropic needs to poke, prod, and guide these models to behave in the desired way does not give the impression of intelligence. It gives the impression of a complicated automaton.

I don't know how people can read stuff like this and think LLMs are intelligent or conscious.

Gave me wrong information on my very first question. Wasn’t even complicated, and I wasn’t trying to trick it.

Are you going to enjoy a future where those different sources can't be found, so now Google requires you to have a subscription that includes data about vehicle repair? The great thing about the web before was that the information was available for everyone, it was decentralized. This is what they are trying to kill.

They want to capture more of the value that was previously going to others. That's basically what this has all been leading to. Why let a cooking website get visitors and ad revenue when they are free to take the content and show it as their own? Now they are going to do the same to e-commerce. Either they are going to let customers buy their products through Google's interface, or they won't be discovered. No more ownership of the customer relationship. Stores will be a backend warehouse and manufacturer now with Google taking a percentage of all profits.

It is just a theory, the conspiracy part is not really applicable. I don't see what is controversial about it. Are you implying the machine taught itself the mathematics to do all this?

How is that a "tin-foil-hat" take? It's not a secret, and in fact widely reported, that these companies are spending billions on creating training data.

I've been thinking about getting the M5 Max Macbook Pro when it comes out to run local models on, but I'm worried it would make the computer sluggish to work with. Is it better to just get some Strix Halo mini pc to be dedicated local machine?

Is this going to be another app that I have to make sure I opt-out of training for?

Do some freelancing to see what problems people will pay you to solve, then figure out if it is a good problem to solve at scale.

Dropshipping is just logistics. There is a lot of dropshipping that has nothing to do with ordering off Alibaba.

Those upsides are currently just a fantasy and ignore the very real current downsides. They also do not in any way rely on AI to become a reality.

It would be difficult, but not necessarily THAT difficult. With enough pushback from the public, AI would start getting regulated in meaningful ways. The problem is too many people love it, and see no problem with it. Because the momentum and money is on their side, it feels like it is impossible. Maybe things will turn out fine and we will just live in a similar but more depressing future, but if the pro-AI crowd gets bit and changes sides that could be a turning point.

That's true, too. I guess we will see if executive pay and credentials start going down. They could technically have AI make all the decisions while someone just plays the patsy.