Meta’s New AI Image Generator Can Make Pictures From Anyone’s Public Instagram Page https://petapixel.com/2026/07/08/metas-new-ai-image-generator-can-make-pictures-from-anyones-public-instagram-page/ #aiimagegenerator #Post-Processing #photoediting #Technology #metaimages #instagram #museimage #aiimage #privacy #metaai #genai #News #meta

Warning, if you missed this yesterday. This practice includes Meta feasting on your photos as well for its dumb AI.

"As part of Meta’s Muse Image model rollout, Instagram users with public accounts need to opt out to block AI generations of their content."

Wired: Meta Now Lets Anyone Use Your Instagram Photos in AI Images—Unless You Opt Out https://www.wired.com/story/meta-now-lets-anyone-use-your-instagram-photos-in-ai-images-unless-you-opt-out/ @WIRED #privacy #Meta #instagram #copyright

I had to do some inventory tasks for #work today at a remote #datacenter. I was listening to #NPR on the drive home and the discussion was on #AI, specifically on how the #tech titans like #OpenAI, #Anthropic, #Google, and #Meta have quietly stepped away from #safety mechanisms as their models improve, the #ethical implications, and #government #regulation. While the quiet distancing from safety is of obvious concern, the discussion on regulation caught my attention.

As one host put it (paraphrasing here) you'd think these companies don't want regulation but that's not necessarily the case. They are looking for regulation that favors them, whether it's getting in bed with the government or squashing out the competition. The bottom line is still about maximizing profits.

The last part was some interesting food for thought and it makes sense. If correct, it's some good indication of where they (and we) are headed.

On the flip side, #ChatGPT and these #LLMs still make mistakes that, unless you have at least *some* previously acquired knowledge on said topic, are believable.

Case in point: A few months ago, prior to one of my lab servers dying, I wanted to host both #Debian and #FreeBSD repos on the box (which was running the former). ChatGPT provided erroneous step by step instructions, including instructions to build FreeBSD Ports on Debian...using apt. 🤦

#news #technews #publicradio

Dark patterns in Mastodon?

Mastodon is not built to extract profit from its users, but is its design is influenced by commercial platforms? There are many parallels between Mastodon and Twitter for example. And there is user pressure to implement features found in commercial platforms - quote toots for example.

I fled to Mastodon when Musk took over Twitter with the explicit goal of making it a transphobic medium, but I arrived on Mastodon with my Twitter addiction intact. I wanted the same hit, and I got it. Now Mastodon is my favored displacement activity as I deal with severe depression. That's not necessarily a bad thing.

The EU recently launched an action against X for its addictive design. One feature they explicitly mentioned was infinite scrolling. I don't see how a microblogging system can avoid infinite scrolling, but ok.

My question is: when Mastodon is designed and implemented, is any thought given to the mental health implications of the new features? It's easy to look at features on commercial platforms and say "hey, we can do that too" but those features on the commercial platforms are designed with hostile intent by teams of malign behavioral scientists.

#mastodon #mentalHealth #meta

Oh, wow! WhatsApp kann jetzt Benutzernamen!

Das ist ja der reine Wahnsinn, ich bin erschüttert: WhatsApp (WA) will jetzt eine Funktion einführen, die Signal schon lange hat, nämlich Benutzernamen. Der Sinn liegt darin, dass ich dank Benutzernamen meine Telefonnummer nicht mehr preis geben muss - jedenfalls meinen Chatpartner/inne/n gegenüber. WA weiß die natürlich trotzdem und kann dank des zusätzlichen Benutzernamens sogar die Dienste-übergreifende Verfolgung noch verfeinern. Die Vorteile und Nachteile (!) des Benutzernamens auf WA hat die VZ NRW aufgedröselt. Wie Anonymität richtig geht, macht schon lange Threema vor. Da braucht man von vornherein keine Mobilnummer, auch

https://www.pc-fluesterer.info/wordpress/2026/07/07/oh-wow-whatsapp-kann-jetzt-benutzernamen/

#chat #identität #messenger #meta #privacy #unplugmeta #UnplugTrump #verfolgung #whatsapp

Look, I want to “protect the children” as much as anyone. I also wish that the cynical, craven executives at Meta would spend their retirements getting by on nothing more than SNAP and social security and Medicaid.

Having said that, the greatest damage Facebook did to the world was mobilising grown-ass adults. Genocide in Myanmar, pyramid schemes and scams that target older folks, so much fraud, stalking and abuse. Fox News-ifying a whole generation of older people. Wrecking news as an industry. Defrauding advertisers. The list goes on and on.

I know that “won’t somebody think of the children?” is politically easy to sell. But it is such a small slice of the harm they’ve done to society. If focusing on the children gets them brought low, fine. But I can’t agree that it’s the most important.

https://gizmodo.com/metas-teen-safety-case-just-became-a-1-4-trillion-existential-threat-2000782306

#meta #facebook

"Thirty-three states have banded together to sue Meta, alleging that the company was exploiting its young users on Instagram and Facebook for profit, including by collecting data from children without parental consent."

The damages could be as high as $1.4 trillion, nearly the same as Meta's market capitalization at just above $1.5 trillion.

https://gizmodo.com/metas-teen-safety-case-just-became-a-1-4-trillion-existential-threat-2000782306

Via https://flipboard.com/@thenewsdesk/technology-shjum1jiz/-/a-CHwwNHlfS-GTffJivjlDbA%3Aa%3A43591897-%2F0

#news #technology #TechNews #meta #facebook #instagram #SocialMedia

Help! Mommy! Why can't I just break things without penalties?

Reuters: Meta says US states are seeking $1.4 trillion in penalties in August youth safety trial https://www.reuters.com/business/meta-says-us-states-are-seeking-14-trillion-penalties-august-youth-safety-trial-2026-07-07/ @Reuters #Meta #Facebook #Instagram #socialmedia

While I really don't have much love for the Mistral products due to being way behind even cheap open weights Chinese models such as the Qwens, DeepSeeks etc of this world, this is an interesting (possibly new) approach:

https://www.heise.de/news/Leanstral-1-5-Mistrals-KI-Modell-fuer-formale-Beweise-ist-Open-Source-11355236.html / https://www.heise.de/en/news/Leanstral-1-5-Mistral-s-AI-model-for-formal-proofs-is-open-source-11355532.html

With the money quote being:
"In addition to mathematical proofs, Mistral demonstrates a pipeline for automatic bug detection in Rust projects. The tool Aeneas translates Rust code to Lean, after which Leanstral derives correctness properties and attempts to prove or disprove them. "

This goes past the very elaborated statistical parrots that LLMs are and goes into the formal verification territory.

As much as I understand this, this is closer to a model checking approach than what we had before and would in theory thus even allow to check for concurrency issues like race conditions etc reliably.

The hard part there of course was to provide the provably correct bridge between the code to be tested and an efficient model checker representation of the same code. Asking the model checker smart "questions" is another part as is translating findings back to the original code.
Since the LLMs guzzle up RAM and CPU already, the consumption of the model checkers / theorem provers isn't an issue anymore (as long as the model they're being fed is efficient...)

Lets see how well the LLMs do on this work.

I'm sure there are already US or Chinese models that do this or will replicate it in a months time.
I think I read about Meta experimenting on this as well.

Did I get the announcement right? Has anyone had their hands on said model+prover combo already? Is anyone aware of Chinese/US equivalent contraptions?
Any updates @heiseonline @heisedeveloper can provide on those questions in the article, especially on competitors?

#llm #ai #formalverification #modelchecker #modelchecking #concurrency #correctness #meta #mistralai #aeneas #leanstral

Gerade bei vielen hier auf Mastodon stößt WhatsApp eher auf Kritik. Wie bei uns – daran ändern auch die neu eingeführten Benutzernamen nichts. Wir haben uns mit dem optionalen Angebot des Messengers beschäftigt: https://www.verbraucherzentrale.nrw/wissen/digitale-welt/apps-und-software/benutzernamen-bei-whatsapp-bieten-nicht-nur-vorteile-123779

#whatsapp #meta #datenschutz #verbraucherzentrale

RE: https://tldr.nettime.org/@remixtures/116867630152249137

#Meta is a stinking pile of garbage.

RE: https://techhub.social/@Techmeme/116856918203941961

Yet another reason to #ban #meta
Horrible.

But the worst part is that Zuckerberg laughs all the way to the Bank

#WhatsApp usernames are already raising impersonation red flags

https://techcrunch.com/2026/07/01/whatsapp-usernames-are-already-raising-impersonation-red-flags/

#privacy #Meta

#Zuckerberg recunoaște dificultăți în strategia #Meta pentru AI, în ciuda investițiilor uriașe.

🔗 https://wp.me/p9KpFA-5nUv

#Știri

🧠#InteligențaArtificială a închis 154.000 de locuri de muncă, în primele șase luni ale anului 2026.
Reduceri masive de personal la #Meta, #Amazon și #Microsoft.

🔗 https://wp.me/p9KpFA-5nRg

#Știri #Tehnologie #AI

#Meta Platforms macht Pläne für KI-#Cloud-Geschäft konkreter

Meta verfügt scheinbar über mehr KI-Cloud-Ressourcen als benötigt. Das Unternehmen könnte nun mit einem Vorstoß gegen Amazon Web Services, Microsoft Azure und Google Cloud beginnen.

https://t.ress.at/bapfS/

Bananas that #Meta admitted that they bought billions of dollars more of #AI compute than it turns out they needed, and their stock went up 9% at the news.

https://youtu.be/okIHWPwRkms?t=67

Onko datakeskuksia jo liikaakin? Mitä ilmeisimmin ei

Googlen kerrotaan joutuneen asettamaan kovia rajoituksia maksavalle, isolle asiakkaalleen, sillä Googlen tarjolla oleva tekoälylaskennan kapasiteetti ei yksinkertaisesti riitä kaikille sitä tarviville.

https://dawn.fi/uutiset/2026/07/02/googlella-on-liian-vahan-datakeskuksia-joutui-rajoittamaan-maksavaa-jattiasiakasta

#google #meta #datakeskus #tekoäly #uutiset #teknologia #tekniikka

https://newsletter.pragmaticengineer.com/p/why-is-meta-destroying-its-engineering

TL;DR
META is crap, and has always been

#meta #crap #exitnow