If you own Meta smart glasses, it's probably time to put them in the bin. Unless, you know, you want all your pictures and conversations swallowed up by Meta's AI leviathan...

https://techcrunch.com/2025/04/30/if-you-own-ray-ban-meta-glasses-you-should-double-check-your-privacy-settings/
https://techcrunch.com/2025/04/30/if-you-own-ray-ban-meta-glasses-you-should-double-check-your-privacy-settings/

Another dark pattern operation involving LLMs/generative AI. Users are rightly upset that they have been manipulated.

(I can't help but be slightly suspicious that Reddit's licensing arrangement with OpenAI colours their legal view a little here, though)

https://www.engadget.com/ai/researchers-secretly-experimented-on-reddit-users-with-ai-generated-comments-194328026.html
https://www.engadget.com/ai/researchers-secretly-experimented-on-reddit-users-with-ai-generated-comments-194328026.html

AI is not coming for your job. Actually, it's probably coming to make your job a whole lot worse.

https://futurism.com/professors-company-ai-agents
https://futurism.com/professors-company-ai-agents

Wow. One jailbreak to rule them all.

https://hiddenlayer.com/innovation-hub/novel-universal-bypass-for-all-major-llms/#The-Policy-Puppetry-Attack
https://hiddenlayer.com/innovation-hub/novel-universal-bypass-for-all-major-llms/#The-Policy-Puppetry-Attack

Coded with "vibes" perhaps? Serious question, if you're a vibe-coding convert, how do you avoid security issues like this? Hard coded secrets is a very basic error.

https://www.forbes.com/sites/rashishrivastava/2025/04/18/perplexitys-android-app-is-infested-with-security-flaws-report-finds/
https://www.forbes.com/sites/rashishrivastava/2025/04/18/perplexitys-android-app-is-infested-with-security-flaws-report-finds/

I've never been one for the gym, but I do believe in maintaining cognitive health through being very careful with what you injest and exercising my brain. Handing off problem solving to AI is junk food for the mind.

https://www.theguardian.com/technology/2025/apr/19/dont-ask-what-ai-can-do-for-us-ask-what-it-is-doing-to-us-are-chatgpt-and-co-harming-human-intelligence
https://www.theguardian.com/technology/2025/apr/19/dont-ask-what-ai-can-do-for-us-ask-what-it-is-doing-to-us-are-chatgpt-and-co-harming-human-intelligence

Reading that the UK government are working on a "murder prediction" algorithm means I dust off this very simple flowchart. New rule: Every government department to have an independently appointed science-fiction author on staff. Read this if you dare: https://tinyurl.com/yv8n8f2b

Back on the horse. Finally "Return to Neverbury" is almost done. 1,140 words, not bad for two hours. #dailywords #amwriting

Another day, another petition. The tide seems to be changing and people are realising the enormous cultural cost of allowing generative AI companies to ride roughshod over copyright law. They can, and must, be brought under control. They can, and must, pay for the content their models are based on. https://chng.it/GV7hsBL45p

Model Context Protocol is proof that agentic AI can't use computers like we do. As predicted, that was a grift. MCP is nothing but a standardisation of APIs to make it *look* like AI uses computers like us.

Bravo Cloudflare. Bravo.
https://www.theregister.com/AMP/2025/03/21/cloudflare_ai_labyrinth/

My birthday is coming up, so I returned the the tree that I used to make my passage into the human realm from my home dimension. Good times.

We need to stop using the term "hallucination" as if these errors are somehow mysterious and beyond comprehension. It's a grift. A con. A scam. And it's dangerous.
https://www.bbc.co.uk/news/articles/c0kgydkr516o
https://www.bbc.co.uk/news/articles/c0kgydkr516o

#gigo

https://www.theregister.com/2025/03/19/llms_buggy_code/
https://www.theregister.com/2025/03/19/llms_buggy_code/

If your search engine being wrong more than sixty percent of the time is a future you're excited about, you have bigger problems than your search engine being wrong more than sixty percent of the time.

https://futurism.com/study-ai-search-wrong
https://futurism.com/study-ai-search-wrong

Say no more.

https://www.pcgamer.com/software/ai/microsoft-co-authored-paper-suggests-the-regular-use-of-gen-ai-can-leave-users-with-a-diminished-skill-for-independent-problem-solving-and-at-least-one-ai-model-seems-to-agree/
https://www.pcgamer.com/software/ai/microsoft-co-authored-paper-suggests-the-regular-use-of-gen-ai-can-leave-users-with-a-diminished-skill-for-independent-problem-solving-and-at-least-one-ai-model-seems-to-agree/

Nvidia must be thanking their lucky stars that the US government will now need a whole load of power guzzling GPUs for crypto, seeing as the bottom is just about ready to fall out of the AI scaling game...

Introducing the boys to Angel and the thing they find most horrific is dial-up internet...

The one thing we seemed to have mastered here is an incredibly efficient way to pour money down the drain. https://search.app/byGVNmUEG2Rx2ZjFA
https://search.app/byGVNmUEG2Rx2ZjFA

This is how the world ends. Not with a bang...