Old Fucking Punk @lwriemen@librem.one

Mexit, not Brexit, is the new priority for the UK - https://www.theregister.com/2025/08/08/opinion_column_mexit_not_brexit/ "A #Microsoft Exit strategy isn’t just a good idea, it’s vital. It must go a long way beyond a farewell to Redmond"

New one

I had the Blueberry talk with GPT5. https://kieranhealy.org/blog/archives/2025/08/07/blueberry-hill/

I keep hearing AI proponents say that it's critical to develop these AI-using skills *now*, because otherwise you'll be left behind.

But isn't the whole point that AI means you can just let all your skills atrophy and let the magic box do things for you?

Any AI system that you can't just sit down and use is *surely* not the True AI. ;-)

“I’m a guy who has been running independent websites and dealing with ad networks for more than 15 years and this book demystified a lot for me.”

https://tedium.co/2025/08/07/ari-paparo-yield-google-antitrust-review/

Elite universities are caving to Trump’s authoritarian demands. Here’s what you can do to fight back: https://indivisible.org/resource/elite-universities-are-caving-trumps-authoritarian-demands-what-you-can-do-next?source=mastodon

Two years ago when researchers found and publicly exposed an intentional backdoor in a TETRA encryption algorithm used to secure radio communications for police/military/intel agencies around the world -- the algorithm involved a key advertised as one strength but secretly reduced to 32 bits -- the European organization that produced the algorithm told users that to secure their communications they could deploy an end-to-end encryption solution on top of the backdoor'd algorithm. Now the same researchers say they found a security problem with the end-to-end solution as well -- another reduced key. Here's my story for Wired:

https://www.wired.com/story/encryption-made-for-police-and-military-radios-may-be-easily-cracked-researchers-find/

Google calendar can be poisoned with invisible, malicious Gemini prompts:

https://www.darkreading.com/cyberattacks-data-breaches/google-gemini-ai-bot-hijacks-smart-homes

Do I have to stop clicking on calendar invitations? What are the alternatives?

If Google can't get security or "AI" right, what hope is there that anyone will?

Get the schools to teach this stuff, and then you'll have compliant people entering the workplace who need subscription for these services to do anything. Brilliant!

"We can blend CS with design, ethics, systems thinking, and human-computer interaction..."

I really can't believe they dropped "ethics" in there with a straight face. Of course, their ethics are probably different from mine. 🤔

We're going to need journalists to stop talking about synthetic text extruding machines as if they have *thoughts* or *stances* that they are *trying* to *communicate*. ChatGPT can't *admit* anything, nor *self-report*. Gah.

https://www.wsj.com/tech/ai/chatgpt-chatbot-psychology-manic-episodes-57452d14

Psst... looking for a new club to join? 👀

Great news, ours is looking for new members: https://codeberg.org

#Australia Completely Loses The Plot, Plans To Ban Kids From Watching #YouTube - https://www.techdirt.com/2025/08/06/australia-completely-loses-the-plot-plans-to-ban-kids-from-watching-youtube/ "The end result will be that Australia has basically taught a generation of teenagers not to trust the government, that their internet regulators are completely out of touch, and that laws are stupid."