I mean, that’s just how it has always worked, this isn’t actually special to AI.

Tom Hanks does the voice for Woody in Toy Story movies, but, his brother Jim Hanks has a very similar voice, but since he isnt Tom Hanks he commands a lower salary.

So many video games and whatnot use Jim’s voice for Woody instead to save a bunch of money, and/or because Tom is typically busy filming movies.

This isn’t an abnormal situation, voice actors constantly have “sound alikes” that impersonate them and get paid literally because they sound similar.

OpenAI clearly did this.

It’s hilarious because normally fans are foaming at the mouth if a studio hires a new actor and they sound even a little bit different than the prior actor, and no one bats an eye at studios efforts to try really hard to find a new actor that sounds as close as possible.

Scarlett declined the offer and now she’s malding that OpenAI went and found some other woman who sounds similar.

Thems the breaks, that’s an incredibly common thing that happens in voice acting across the board in video games, tv shows, movies, you name it.

OpenAI almost certainly would have won the court case if they were able to produce who they actually hired and said person could demo that their voice sounds the same as Gippity’s.

If they did that, Scarlett wouldn’t have a leg to stand on in court, she cant sue someone for having a similar voice to her, lol.

This seems like it has pretty powerful potential for space flight.

Being able to aggressively min max packaging materials to secure materials could be critical for reducing payload sizes on shuttles, where every single individual gram counts.

Each kg of packaging is thousands of dollars to get into orbit, so that’s really appealing.

I’d be curious to see if Amazon is also working on box packing algorithms for maximizing fitting n parcels across x delivery trucks.

IE if you have 10,00 boxes to move, what’s the fewest delivery trucks you can fit those boxes into as fast as possible too, which introduces multiple complex concepts. Both packing to maximize space usage and the order you pack it in to minimize armature travel time…

I’d put money down amazon is perfecting this algorithm right now, and has been for awhile.

Bruh did you read the article at all? Nothing you talked about has anything to do with what this AI is for.

Sometimes its a physical issue in your setup.

Double check your cable, double check the carriage, and double check the rails, look for potential obstructions.

I had one print that kept failing in the exact same place each time, couldn’t figure it out, then I watched it live and the dang ribbon itself was physically catching on a specific part of the geometry mid print and then the print would twist a bit, lol.

Something to consider, I’d recommend visually watching that specific layer when it’s coming up to see if you see something happen.

I primarily use a couple locally hosted LLMs for searching for info now.

Larger LLMs are trained on so much info that they get the answer right surprisingly often.

Only thing they of course struggle with are recent events.

I’ve been calling this for awhile now.

I’ve been calling it the Ouroboros effect.

There’s even bigger parts at play the paper didn’t even dig into, and that’s selective bias dye to human intervention.

See at first let’s say an AI has 100 unique outputs for a given prompt.

However, humans will favor let’s say half of em. Humans will naturally regenerate a couple times and pick their preferred “cream of the crop” result.

This will then ouroboros for an iteration.

Now the next iteration only has say 50 unique responses, as half of them have been ouroboros’d away by humans picking the one they like more.

Repeat, each time “half-lifing” the originality.

Over time, everything will get more abd more sameish. Models will degrade on originality as everything muddles into corporate speak.

You know how every corporate website uses the same useless “doesn’t mean anything” jargon string of words, to say a lot without actually saying anything?

That’s how AI is going to local minima to, as it keeps getting selectively “bred” to speak in an appealing and nonspecific way for the majority of online content.

Not related to the article at all mate.

This article is about how many plugins have Bern discovered to have implemented oath in a very insecure way and simply using them can expose your sensitive info you have linked to your chatgpt account.

IE:

1. You connect your github account to your chatgpt account (so you can ask chatgpt questions about your private codebase)

2. You install and use one of many other compromisable weakly implemented plugins

3. Attacker uses the weak plugin to compromise your whole account and can now access anything you attached to your account, IE they can now access your private git repos you hooked up in step 1…

Most of the attack vectors involve a basic (hard to notice) phish attack on weak oath urls.

The tricky part is the urls truly are and look legit. It isn’t a fake url, it actually links to the legit page, but they added some query params (the part after the ? In the url) that compromise the way it behaves

Note that ChatGPT indeed implemented a state parameter, but their state was not a random value, and therefore could be guessed by the attacker.

Bruh wut, rookie mistake.

State is supposed to be mathematically random and should expire fairly quickly.

I always have used a random guid that expires after 10-15 minutes for state, if they try and complete the oauth with an expired state value I reject ad ask them to try again.

Also yeah the redirect uri trick is common, that’s why oath apis must always have a “whitelist urls” functionality. And not just domain, the whole url.

That’s why when you make a Google api token you gotta specify what urls it’s valid for explicitly. That way any other different redirect uri gets rejected, to prevent an injection attack from a third party providing their own different redirect uri to a victim.

Oath is pretty explicit about all these things in its spec. It really sucks people treat it as optional “not important” factors.

It’s important. Do it. Always.

Oh geez, my bad, I thought it was Google.

I trust OpenAIs demos more for largely the reason you mentioned, they have tended to be more transparent about the downsides of their products before.

Fair enough!

Ah yes, Sora, put out by Google, who definitely hasn’t been known to have already done a bunch of very misrepresentative footage on their past AI demos…

Something tells me when push comes to shove it’ll become obvious that though Sora will be cool and revolutionary, it also won’t be nearly as big of a deal as they made it look in the demo and folks will notice efforts were made to selectively cherry pick shots that heavily favor the tool, and as soon as you go outside its wheelhouse it starts to struggle.

Edit: nevermimd I was very much mistaken!

Often these types of articles tend to be very “a lot written but nothing said”, but that is very much not the case for this article.

I really enjoyed how in depth it went both on the history of how things used to be compared to what the new advancements will provide.

Outsourcing through the foundry program might lead to some pretty big revolutions in chip making. I wonder if we will start to see open source chips start to show up as large companies open the floodgates to let individuals contribute to make improvements on dies…

Which makes me question whether the fediverse is a good thing

I’d argue it’s good, because it means open source AI has a fighting chance with FOSS data to train on without needing to fork over a morbillion dollars to Reddits owners.

Whatever use cases the reddit data can train on, FOSS researchers can repeat it on Lemmy data and release free models that average joes can use on their own without having to subscribe to shit like Microsoft Copilot and friends to stay relevant.

They definitely do, it’s common for such systems to never actually delete anything because storage is cheap. It likely just is flagged deleted=true and the searches just return WHERE [post].Deleted = False on queries on the backend.

So it looks deleted to the consumer, but it’s all saved and squirreled away on the backend.

It’s good to keep all this shit for both legal reasons (if someone posts illegal stuff then deletes it, you still can give it to the feds), as well as auditing (mods can’t just delete stuff to cover it up, the original still exists and admins can see it)

1. Called this awhile back, this is why Reddit has such a high evaluation.

2. Poisoning your data won’t do anything but give them more data, do you seriously think reddit servers don’t track every edit you make to posts? You’d literally just be providing training data of original human vs poisoned. They’d still have your original post, and they have a copy of everytime you edit it.

3. Whoever buys reddit will have sole access to one of the larger (I don’t think largest though) pools of text training Data on the internet, with full licensed usage of it. I expect someone like Google, FB, MS, OpenAI, etc would pay big $$$ for that.

“But can’t people already scrape it?”

1. Well yes, but it’s at best legally dubious in some places

2. Scraping Data off reddit only gets you current versions of posts (which means you can get poisoned dara, and cant see deleted content), and is extremely slow… if you own the server you have first class access to all posts in a database, including g the originals and diffs of everytime soneone edited a post, and all the deleted posts too.

Think about if you perhaps wanted to train an AI to detect posts that require flagging for moderation, if you scrape reddit data, you can’t find deleted posts that got moderated…

But, if you have the raw original data, you 100% would have a list of every post that got deleted by mods and even the mod message on why it was deleted

You surely can see the value of such data, that only owners of reddit are currently privy to atm…

Im curious to see what sorts of recommended minimum specs there will be for these features. It is my understanding that these sorts of models require a non negligible amount of horsepower to run in a timely manner.

At the moment I am running Nextcloud on some raspberry pis and, my gut tells me I might need a bit more oomph than that to handle this sort of real time AI prompting >_>;

Yes, its largely inexpensive to spin up a new batch of collectibles, so the vast majority of them are indeed worthless because anyone can just make one.

The small handful that are actually collected by the public at large and are considered worth something… continue to be worth something.

It’s like grouping thousands of meaningless mass produced collectable trading cards made by game companies that no one actually gives a shit about, in with the likes of MTG/Pokemon/Hockey/Baseball/etc

Like yeah if you look at it as a whole, no one gives a shit about 99% of them, because its literally just ink on paper, so its not exactly that hard to shit out yet another “collectible” no one cares about.

But the public will latch onto the very small handful that actually hold some form of appeal, and those will absolutely continue to hold value.

99.9% of Magic the Gathering cards aren’t worth the ink and paper they are made of.

Doesnt have any bearing on whether your mint condition Black Lotus is worth $$$$

Say what you want, but its also an iconic look. You recognize his face the moment you see it, with that sorta disinct (but still kinda shit looking) haircut.

But also I dunno, I can understand the vibe of wanting super short hair to keep it up and out of the way, some people just prefer their hair cut to stay the hell out of the way, without committing to being full shaved down.

I try not to judge on such a thing. If he likes that haircut, fuck it, the dude makes more money while sitting down for his haircut then 100 haircuts would cost him to pay for.

He can do whatever the fuck he wants with his hair.

I would rather judge him by his actions, like that shit he pulled with buying “private” beaches and whatnot.

Edit: Actually out of curiosity I looked this up for some updates, and it sounds like he and his wife have been putting in efforts to actually do right by the locals, and that a lot of the bad press was just trying to dunk on him and was largely just stuff going over poorly with locals, so he retracted offers and went back at it again but with the help of local professors to ensure he did it right and in a better way.

It sounds like over the past few years he and his wife have been pouring money, like a LOT of money, into preserving tonnes of wildlife and donating a bunch of money to save large areas that were going to be privately developed, and instead now have enough money to pretty much stay preserved and managed by the locals forever.

https://www.kauaitravelblog.com/mark-zuckerberg-kauai-property/

Not gonna lie, it actually sounds like the dude is trying to put in some real work to do good here, I have to say.

Yes, of course the actual autistic people would know less about how to address their daily issues than doctors /s

Its been shown time and time again though that the people who are gaining attention/views/money on tiktok and whatnot… are not exactly likely to be telling the truth.

People figure out very fast whatever magic flavorful words they need to say every month to farm the clicks and get those likes and shares.

I’d expect the majority of people you see claiming they are autistic on tiktok and proceeding to start giving medical advice about it, are likely just lying for money.

It becomes even more obvious when you look at their history and see that what they focus on shifts every few months in terms of content they push.

Step 0: Be the sort of scum that would assault a lyft driver
Step 1: Set your identity to non binary as a rider
Step 2: Dress in a way to appear non-binary, even a little bit. Honestly just painting your nails purple and wearing thick glasses is probably enough to not raise suspicion. Most people wont try and question this and interrogate you over it. If they do, filter them off and be a normal rider.
Step 3: If they don’t question it, congrats, Lyft has no just done the work of assisting you with finding your next victim, great job Lyft!

Bonus round~!

1. Be a nazi
2. Do steps 1-3 above
3. Set your destination to be somewhere vaguely secluded where your fellow nazi friends are lying in wait.
4. Congrats, Lyft has now successfully routed a non-binary identifying person directly into you and your nazi friend’s clutches, great job lyft!

Heyo!

Query: Have you tried out Kavita yet? It’s what I have started for self hosting my books/PDFs/etc, and there’s a lot of free open license books out there that you can bulk load onto it and have a whole library at your fingertips.

I believe there’s a way to integrate it with kindle as well?