@clarablackink Agree all the way. I really appreciate Timnit Gebru’s work on TESCREAL, with regards to eugenics. I try to bring this point forward in conversations sometimes but am often met with skepticism. People don’t like to hear that the tools they are using enable racist thinking.
Tagged: Akkoma
@robertklaschka Ah, nice! Thanks for sharing.
@tymwol Yeah, I should stay away.
@tymwol Yeah, I should stay away.
I was reading a debate on LinkedIn where one side was arguing that "AI" image generators are more energy-efficient than hiring a photographer (or buying a stock photo).
One argument that made me do a painful face-palm was the notion that photographers need to eat, so they consume more energy.
The thing about humans is… wait for it… they ALWAYS need to eat. You can’t shift production to a computer and argue that it’s more efficient because it doesn’t eat. The human will keep eating(!) You are only adding energy-use.
These types of arguments illustrate a deep obliviousness towards human wellbeing. The argument only works if you expect the human to eat less when their job is taken away. Or die.
Their only data point for “AI" energy use was the point of generation, neatly excluding:
- all the energy used when scraping the internet for all the required datasets
- the energy used when building the model itself
- the energy used for required human workforce tasks, such as labeling (I bet those humans have to eat too)
- the energy used when building required data centers, both in-use and speculative investments (so many eating humans involved)
- the energy for hardware production and ongoing replacements
- the fact that the ease-of-use (and being “free”) encourages way, way more image generation than ever before…
Oh, and one more thing…
The image generator feeds on, and requires, photos and images taken and made by humans… who eat. Damn those humans.
Oh wait, the person doing the prompting… that person has to eat as well. It’s almost like it’s a human imperative.
Worth noting is also that none of the big "AI” companies are at all transparent about their energy consumption for this technology. Nor water consumption. Nor noise pollution. Nor e-waste. So there’s a ton of speculation.
I just wanted to make clear that I’m all for allowing humans to eat.
I know it's relevant to refer to GIGO (garbage in, garbage out) when talking about language models.
But it's also important to consider FIGO - faith in, garbage out - because faith will have a huge impact on the willingness to trust the garbage.
We already have laws that allow us to prosecute people who oppress, enslave, harm, and kill other people. The challenge is to make visible how it happens even when the method used is "innovative technology”. That’s when we can hold people accountable, and stop blaming the technology.
Machines don’t hurt people. People hurt other people. With machines. And sometimes the crime happens ever so slowly across time, space, borders and multiple touchpoints. Which is why it can be hard to note the similarity to a whip or a bullet.
@abbe98 Very much agree. I’ve also been doing thought experiments along those lines… as in how we could develop visualisations of big and heavy files, or network congestions.
@ewen @johnquiggin
I do love that article by @baldur
@AngelaScholder
Yes, it’s a fork of Pleroma. Note that it’s continuously update but there is not really any support given for it so you kinda need to have your own network of people you can reach out to if you run into trouble. I managed to install it with below average server management skills though. 😅
More info: https://joinfediverse.wiki/Akkoma
@abbe98
@abbe98
I hear you and I’m not sure I mean the act itself is what makes things better, but rather that the awareness of costs makes you behave in a completely different way. Since actions in a self-hosting environment are felt more I believe more care is taken, which has potential domino effects for related behaviors.
It’s one thing to consider the difference between cloud-storing or self-hosting an image, another thing to choose not to store the image online at all.
@abbe98 It’s definitely an interesting point. One primary reason I switched from Mastodon to Akkoma was how bloated I felt the software was.
When you’re running all your websites and apps on your own server, and traffic and space costs real money, you are way more careful with optimising images and files, and refrain from unnecessary uploads.
When people use “free" services, paying with their personal data and biometrics, they don’t seem to care.
This is one example of how free services are bad not just for privacy, but also for the environment – because all those costs are hidden.
I just realised that in two years time I will have been blogging on my own domain for 30 years… that’s a party I should start planning now.
Enkelt och bra av Emma Frans idag.
Jag älskar dock inte idén om att modellen ”bullshittar” heller, vilket också frammanar bilder av något med egen vilja.
ChatGPT är en köttkvarn för text och bild med en påkopplad sannolikhetsspridare – som inte kan sluta mala.
https://per.ax/efai
@datarama Cool, thanks for sharing! And for alerting me to Runbox. It looks like I should add it to my suggestions here
https://axbom.com/big-trek/
Meta’s AI copyright battle.
"The social media giant is not denying that it used the work or that it downloaded books from shadow libraries en masse, but insists that its behavior is shielded by the “fair use” doctrine, an exception in US copyright law that allows for permissionless use of copyrighted work in certain cases, including parody, teaching, and news reporting.”
US District Court Judge Vince Chhabria: “If you are dramatically changing, you might even say obliterating, the market for that person's work, and you're saying that you don't even have to pay a license to that person to use their work to create the product that's destroying the market for their work—I just don't understand how that can be fair use”.
Reporting by Kate Knibbs:
https://www.wired.com/story/meta-lawsuit-copyright-hearing-artificial-intelligence/
In June, to be finalised in July at the latest, I will be moving my email accounts from Google Workspace (prev. Google Apps) to Migadu.
Likely I should document all this along the way.
Here's the brutally honest and readworthy Pro/Cons list for Migadu:
https://www.migadu.com/procon/
Yes I'm embarrassed to still be dependent on Google for this, but that's also why I'm happy this is finally happening.
”I made this in no time at all!”
”Did you go through and validate the coherence, the messaging and the explanatory approach? Not to mention the accuracy?”
”Nah, man, the point is to save time.”
Thinking as little as possible is now a badge of honor.
Living in the upside-down.