Meta to chociaż poinformowała, że będzie wykorzystywała dane ludzi do trenowania #AI.

#X, który też z nich korzysta do trenowania chatbota #Grok, nie uznał za stosowne nikogo o tym informować.

Do organów ochrony danych osobowych w 9 krajach trafiły skargi przygotowane przez austriacką organizację @noyb.eu - na bezprawne wykorzystywanie danych osobowych użytkowników w UE/EOG do szkolenia technologii X.

Jedna skarga jest z Polski (a my maczaliśmy w tym palce)!

Źródło: https://noyb.eu/en/twitters-ai-plans-hit-9-more-gdpr-complaints

Need a brief and accessible explainer about the actual harms of AI?

Show folks this:

https://m.youtube.com/watch?v=TtVJ4JDM7eM

@thejuicemedia hitting it out of the park, yet again.

#ai #corporations #corporatocracy #billionaires #climateCrisis #BigTech #SiliconValley

Honest Government Ad | AI

The Government™ has made an ad about the existential threat that AI poses to humanity, and it’s surprisingly honest and informative👉 Help us keep Government...

^YouTube

CC @fedora (https://fedoraproject.limequery.com/142117?lang=en)

#GenerativeAI #AI #LLM #ethics #fedora #openSource #FOSS #BigTech #RedHat #IBM https://det.social/@jlink/112811798901196501

#AaronSwartz killed himself because he scraped research data and they aggressively prosecuted him.

Now #AI companies say they should scrape our data for free.

Oh, but it will do that. Reasonably soon.

#computer #science #AI #humor

Image-scraping Midjourney bans rival AI firm for scraping images | Ars Technica

https://arstechnica.com/information-technology/2024/03/in-ironic-twist-midjourney-bans-rival-ai-firm-employees-for-scraping-its-image-data/

Or the rivality between burglars...
#AI

Image-scraping Midjourney bans rival AI firm for scraping images

Midjourney pins blame for 24-hour outage on "bot-net like" activity from Stability AI employee.

^{Ars Technica}

Ed Zitron comments in the last episode of tech won't save us something that I find interesting to point out about #generativeAI , even if it is obvious:

Let's imagine we all accept #AI generators and create a new internet based on them (or a large majority, at least). In that case (which is somewhat the business model pursued by #bigtech), we would reach a point where systems would be able to be trained only on content created by themselves.

Thanks @parismarx 🥰

https://www.techwontsave.us/episode/211_silicon_valley_deserves_your_anger_w_ed_zitron

Silicon Valley Deserves Your Anger w/ Ed Zitron - Tech Won’t Save Us

A left-wing podcast for better technology and a better world.

^{Tech Won't Save Us}

𝟮𝟬𝟭𝟬𝘀…
I worked hard on some 3D artwork. Now I'm going to publish it on social media and enjoy the likes and responses. 😊

𝟮𝟬𝟮𝟬𝘀…
I worked hard on some 3D artwork. Before it will be ready to publish, I need to make a compilation of work-in-progress screenshots and wireframes, to prove that it's not AI-generated, and to avoid people asking what AI prompt I used to generate the image. 😖😖

#AI #ArtificialIntelligence #tech #BigTech #GenerativeAI #design #art #3D #CreativeToots #FediArt #MastoArt

My company recently internally promoted using Microsoft Copilot for Web including to "Generate creative content (such as poems, stories, songs, and more).". I replied to the Slack message some snarky questions about author consent to have their work used by the LLM tool, after which I was directed to an internal feedback form about LLM usage.

I repeated my question "Do we have the details on the dataset Copilot was trained on and if the relevant authors were asked for their work to be used, especially for the “creative content generation” part?" and I received a personal response from the Senior Vice President of End User Technology & Experience that "No data is tracked, sent to LLM models and microsoft doesnt (sic) know what we ask and return".

I was initially only venting, but the escalation culminating in a non sequitur has made me even more frustrated.

#LLM #AI

Just as we all knew all along, #Reddit’s move to charge exorbitant fees for API access was so they can cut lucrative deals to sell user-created material to train #AI models.

I was a redditor of about 15 years when I decided to delete everything I wrote and close my account because I foresaw this hostile repurposing of what is essentially a massive store of freely-accessible, context-rich community-generated content, made by people who didn’t necessarily want their content to be used for training without compensation.

AI training is largely built on top of exploitation of labor for private gain. It’s legal in many cases, but it’s often unethical and immoral.

“Reddit Signs AI Content Licensing Deal Ahead of IPO”

https://www.bloomberg.com/news/articles/2024-02-16/reddit-is-said-to-sign-ai-content-licensing-deal-ahead-of-ipo

No-paywall version here: https://archive.ph/caW1Y

Reddit Is Said to Sign AI Content Licensing Deal Ahead of IPO

Reddit Inc. has signed a contract allowing a company to train its artificial intelligence models on the social media platform’s content, according to people familiar with the matter, as it nears the potential launch of its long-awaited initial public…

^{Amy Or (Bloomberg)}

My problem with #AI isn’t that it’s a new medium that would somehow threaten the existing ones, like printed books, radio, cinema, television, video games, social media and live streaming did before.

No, my problem is that it’s exactly the same mediums we’re used to, churning an endless stream of a cultural average, reifying it in the process, all the while infringing copyright on an industrial scale and guzzling energy and water like there’s no tomorrow.

It’s the exact opposite of a new medium that grants a new creative freedom and rewards trailblazing artists, instead flattening entire aesthetics genres into their most pleasing surface-level aspect thanks to the absence of creative effort to produce facsimiles at scale.

FB just offered me an ad titled "Learn ChatGPT in just 4 weeks!"

I have so many questions.

Like, is the course taught by ChatGPT? Is the material about ChatGPT written by ChatGPT? Does it come with a certificate? Is there an exam? How do I know the material written by ChatGPT about ChatGPT is accurate?... Does it matter?...

#WTF #WhatTheHell #Why #How #Who #Why #AI #ChatGPT

#AI #LLM is gunking up the web, especially for lesser-represented languages. Spammers are creating garbage English language content using LLMs, then translating it into *multiple languages* at the same time, using Machine Translation, presumably to generate clickbait ad revenue in several languages at once.

In English, such gunk accounts for some 9% of total sampled web content. But in languages with less representation on the Internet, the figures could be much higher. In Malay, it’s something like 26%, and in Swahili it’s nearly HALF of everything found on the web.

Paper [pdf]: “A Shocking Amount of the Web is Machine Translated: Insights from Multi-Way Parallelism”

https://arxiv.org/pdf/2401.05749.pdf

We’re over a year into this cycle of AI hype, but how does the real impact of the technology compare to what tech CEOs have spent all their time warning us about?

On #TechWontSaveUs, I spoke to @timnitGebru about how they distracted us from the real problems with AI to shape regulation and serve themselves.

https://techwontsave.us/episode/203_ai_hype_distracted_us_from_real_problems_w_timnit_gebru

#ai #openai #samaltman #tech

Apologizing for using AI-generated image looks like the new sport for public relations managers in 2024 🤣 :

- Wacom: https://nitter.net/wacom/status/1744814055137919388#m

- Magic: The Gathering: https://nitter.net/wizards_magic/status/1744056808254173447#m

Src on this discussion: https://mastodon.gamedev.place/@lritter/111730964394827333

#ai #wacom #magicthegathering #pr

Leonard Ritter

2024-01-10 09:33:37

<a few weeks ago>

PR guy at wacom: Fuck. FUCK FUCK FUCK! This whole AI shit is worrying budding artists so much, they're giving up on their careers! Our tablet sales are tanking! This AI craze is doing nothing for us! What a lemon! Unless... hmm! What if if we used it to generate ad illustrations? We could cut costs and at least squeeze *some* lemonade from this...

<a little bit later>

PR guy: oh no

So, CEOs...

#Capitalism #AI #work

What are the big takeaways from 2023? @xoxogossipgita, @molly0xfff, and Aaron Thorpe join me for a great year-end chat on AI hype, tech villains, and other big stories of the year!

Listen to the full episode: https://techwontsave.us/episode/200_the_year_in_tech_2023_w_gita_jackson_molly_white__aaron_thorpe

#tech #2023 #elonmusk #jeffbezos #ai

Open source #AI has been a key issue in policy debates, and #AIAct includes provisions that regulate development and sharing of #opensource AI development.
We've been following the proposed rules as they meandered through several different approaches. Now elements of the final version have been made public.

The big question was whether transparency and other obligations need to be mandated for open source AI, or they can be self-regulated – under the assumption that open source developers ensure these elements based on principles of open development.

The agreed upon wording of the AI Act assumes the later, and makes open source exempt from regulation of general purpose AI models, including transparency obligations.

We think that this is a problem, especially that lack of agreed standards that define open source AI means that there is an risk of open washing.

You can read more about this on our blog: Paul Keller wrote a detailed analysis of the provisions:
https://openfuture.eu/blog/a-frankenstein-like-approach-open-source-in-the-ai-act/

A Frankenstein-like approach: open source in the AI act – Open Future

Late last week, the European Commission, the Member States, and the European Parliament reached a deal on the AI Act.

^{Open Future}

I’m somewhat relieved I can still spot a machine-generated product page leading to a very-likely-non-machine-generated checkout funnel. I firmly believe this ability won’t last long, though.

And I’m really curious what you would end up getting in the mail if any.

#AI #ArtificialIntelligence

🔥Clearance Sale🐱Art design Cat Chair

^{Householcable}

Not blown away by Google #Bard's sophisticated reasoning capabilities just yet.

#AI

Check this out:

Your AI-free Content Deserves a Badge

https://notbyai.fyi/

⚠️ Edit: Please read the license and the following toot ⚠️

https://floss.social/@amin@alpha.polymaths.social/111461825808009435

It is not ideal. If I had known this and read the fine-print, I probably wouldn't have shared it 😔

#AIFree #AI #ArtificialIntelligence #NotByAI #LLM #LLMs #SALAMI

Not By AI — Add the Badge to Your Human-Created Content

Download and add the Not By AI badge to showcase your AI-free & human-first approach to content creation (writings, art, photos, paintings, videos, and music)

^notbyai.fyi

Can somebody please get an AI image generator to generate a fox with a human face or head for me?

The other way around is really easy and does not count.

#AI #StableDiffusion #DALLE #Midjourney

Now this is interesting!
"This new data poisoning tool lets artists fight back against generative AI"
https://www.technologyreview.com/2023/10/23/1082189/data-poisoning-artists-fight-generative-ai

#ai #ImageGenerator #ArtistsRights #generativeAI #aiModels #DataPoisoning

This new data poisoning tool lets artists fight back against generative AI

The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models. 

^{Melissa Heikkilä (MIT Technology Review)}