Chris Lynch

AI. As popular as a papercut to the eyeball.
Please support real, organic writers. Some of us are actually quite nice, when you get to know us.

Tonight's #soundclash was an ABSOLUTE BELTER. Wall to wall rock classics. Grab the podcast tomorrow or catch the repeat on the weekend. Whatever you do, do not miss this episode. (It's the best one we've done since the infamous "dinosaur erotica" episode)

31 days to go. Return to Neverbury with me this Halloween for magicians, dodgy lawyers, creeping horrors, changelings, folk music purists, portals to hell, idle gods, ghosts, militant amateur cycling clubs, and a serious problem with something called “Big Chicken”.

Introducing the GenAI quick-step:

1. Create slop machine
2. Do nothing to discourage slop
3. Add your own slop into feeds (see Vibes)
4. Add in adverts (see OpenAI adding adverts and new feeds)
5. Charge a subscription to clean the feed of all the junk you created
6. Profit?

This certainly explains a lot... #AI #lies

For anyone confused... #read

Klarna's CEO says he vibe codes to save his engineers and product managers time Klarna CEO Sebastian Siemiatkowski said AI coding tools like Cursor have helped him build prototypes faster, even with a non-technical background.

This feels a lot like when a disruptive kid is given a toy to play with whilst the adults get on with the work...
share.google/i2jWm4GfVR67...

IP Theft = Theft.

The struggle is real. #restingwitchface

Last of these. Honest. If you're a #doctorwho fan in the #peterborough area and you didn't go to the Adventures in Time and Space exhibit at the Peterborough Museum & Art Gallery, then you have until November 2nd to get down there or invent time travel for yourself.

Ah, I'd love to be better than this, but I'm not. Look! It's me! And a book I wrote! In a shop! (And I didn't have to bribe them or sneak it in myself)

Called it. Nothing more toxic than "The Touch of Zuck".

Shout to my guy, Ham the Astrochimp.

At last... the mystery has been solved.

Can we stop now?

Trouble in paradise?

Why Eminem is taking on Meta in $109 million case Please refresh the page or navigate to another page on the site to be automatically logged inPlease refresh your browser to be logged in

Zuckerberg Vs Eminem? I know who I've got my money on.

www.independent.co.uk/bulletin/cul...

We are facing a future where Google kills the open web as we know it. Why wouldn't content creators hide their content behind paywall and registrations to protect their IP for avaricious AI companies trawling every last byte of data they can?

Microsoft Copilot is an absolute buffoon. When I'm writing an email, in it's desperation to inject itself into my workflow, it keeps suggesting that I sign my emails "Best wishes, Christmas".

Christmas. As I live and breath.

Anthropic faces backlash to Claude 4 Opus behavior that contacts authorities, press if it thinks you’re doing something ‘egregiously immoral’ Bowman later edited his tweet and the following one in a thread to read as follows, but it still didn't convince the naysayers.

Reason 635 why you should not give an AI/LLM access to your data or systems.

venturebeat.com/ai/anthropic...

Anthropic blames Claude AI for ‘embarrassing and unintentional mistake’ in legal filing Well that’s not a good look.

AI firm includes AI hallucination in own legal filing. Physician... heal thyself.

www.theverge.com/news/668315/...

And that's game. 5584 words in 1/2 a day. #dailywords #amwriting That'll do 🐖

4198 words in and lunch has arrived courtesy of Mrs. L, my muse. (And maker of the best lunches in this or any other universe)

Today is the day I'm due to send in my next Havoc Files story for the good folks at Candy Jae. So, in best Chris fashion, I got up this morning to edit, decided everything I've done is wrong, and have started again. 7K words in one day? Absolutely fine. What can go wrong?

If you own Ray-Ban Meta glasses, you should double-check your privacy settings | TechCrunch Meta has updated the privacy policy for its AI glasses, Ray-Ban Meta, giving the tech giant more power over what data it can store and use to train its AI models.

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...

techcrunch.com/2025/04/30/i...

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)

www.engadget.com/ai/researche...

An Entire Company Was Staffed With AI Agents and You'll Never Guess What Happened An experiment by researchers at Carnegie Melon University staffed a fake software company with AI Agents, and the results were dismal.

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

futurism.com/professors-c...

Novel Universal Bypass for All Major LLMs HiddenLayer’s latest research uncovers a universal prompt injection bypass impacting GPT-4, Claude, Gemini, and more, exposing major LLM security gaps.

Wow. One jailbreak to rule them all.

hiddenlayer.com/innovation-h...