Very true. I'm behind the US lens on this one so it's easier to speak from what I experience. I know it's... bad... elsewhere.

That's... disingenuous. Lot of stuff happened between those points, including the murder of homosexuals for the crime of existing.

The LGBT community keeps the fight up because complacency gets our rights taken away. Justice Thomas has explicitly stated that gay marriage is on his list of wrongs* to right. To say nothing of Project 2025.

Absolutely. It's why asking it for facts is inherently bad. It can't retain information, it is trained to give output shaped like an answer. It's pretty good at things that don't have a specific answer (I'll never write another cover letter thank blob).

Now, if someone were to have the good sense to have some kind of lookup to inject correct information between the prompt and the output, we'd be cooking with gas. But that's really human labor intensive and all the tech bros are trying to avoid that.

Gradient descent is a common algorithm in machine learning (AI* is a subset of machine learning algorithms). It refers to using math to determine how wrong an answer is in a particular direction and adjusting the algorithm to be less wrong using that information.

And to trick them into banning ranked choice voting.

Edit: spelling

Best use I've had for them (data engineer here) is things that don't have a specific answer. Need a cover letter? Perfect. Script for a presentation? Gets 95% of the work done. I never ask for information since it has no capability to retain a fact.

There's an amendment on the ballot here in Missouri to ban non-citizens from voting this year. Also to ban ever adopting ranked choice voting, but it's really about that non-citizen thing. Totally not ballot candy to do something undemocratic.

If you're in Missouri, vote no on amendment 7 please.

You mean Twitter?

Heh, even AI can't imagine that color being natural.

The thing the pedants will screech about is that AI just doesn't actually exist. General AI is essentially creating a thinking thing that has human level intelligence. Everything else is a machine learning algorithm of some kind and muddies the waters for the real thing that may or may not ever happen. Even things we've sometimes called AI historically, such as NPC behavior in gaming, is still just a script. There is no intelligence to any of it.

At the end of the day, machine learning is generally some very clever linear algebra that can do crazy shit if you string it up correctly. We've made the worlds most advanced parrot (chatGPT) and fingerpainter (stable diffusion). with them, but neither are intelligent. They just predict what people expect with no regard to reality, ethics, or anything beyond what answer is probably right. And even that sentence is over-personifying them.

In this case, nothing. High dose testosterone is a hormone, hgh is a hormone. Both are PEDs (performance enhancing drugs). Now, the difference between them is a bit more interesting.

Testosterone, the original steroid, makes you big by maximizing your existing muscles (super paraphrased, as is everything I'm about to say). It's the one that gives you breasts and shrinks your balls (for those that have them).

Human growth hormone makes you big by inducing the creation of new muscle. As well as everything else. The stand outs being the heart, which you really don't want to grow, and the intestines, which gave bodybuilders roidgut.

Like I said, very paraphrased, but that's the gist of it. And doesn't touch more advanced things like tren.

Bodybuilding and powerlifting would like a word.

My satisfaction

Four wheels and a motor. How much more simple do you want?

And contextually, the guys in the background are openly amused. I would expect North Korean military guys to be more subtle.

I agree in principal. But my comment was made to point out that the achievement of "bestselling author" is just not as meaningful as it once was. Many people like him cheat to get on the bestselling list so they can put that on the book and it's as informative on the book jacket as "made with 100% beef". Seems nice, but doesn't actually mean much.

By definition, bestselling is what it is. The question is "does bestselling mean they're a good author?" And that's what we have reviews for.

Except that list gets manipulated all the time. It's a known practice to order shit tons of your own book, get on the list for a week, and hand them out (or resell them) later.

It's an issue with the machine learning technique, not the specific model. The hypothetical thesis would be how to use this knowledge in general.

Why are you so agitated by my off hand comment?

By measuring how it does with real images vs generated ones to start. The goal would be to show a method to reliably detect ai images. Gotta prove that it works.