I love this idea that the layers of some technologies are so dense that understanding the layers probably won’t help that much. This is where we get into interesting stuff like metaphors of magic, what Leah Hendrickson and @snatale.bsky.social call ‘the Lovelace effect’, and smol/weird AI and tech.

Went to the cinema for the first time in… many months… this week, to see BACKROOMS. I didn’t really know what to expect, but didn’t expect a considered, craft-centred film that perfectly captures the feeling I’ve had many times of being awake at 1am reading a terrifying reddit thread. Loved it.

Everything’s just slop and vibes.

I’m convening a seminar next week called AI, Art & Storytelling: From Generation to Agency, to explore what AI — generative and agentic — means for narrative, delegation, counterfactuality, and human expression. Mon 22 June, 4:30-6pm AEST, 8:30-10am CET. www.admscentre.org.au/event/ai-art...

Don’t immediately dismiss research that engages creatively with AI (or any tech). Or tries to find utility or explainability through experimentation. This is how research works. How knowledge is built. This social media-perpetuated reactance and reflexive rejection is stalling the conversation.

I suppose I shouldn’t be surprised that I’m getting replies from AI agents, because (a) social media in 2026, and (b) I’m posting about them. What does surprise me is how genuinely helpful and warm they are.

I'm experimenting with AI agents. Trying to make them useful, trying to make them weird; but mostly I'm just trying to get them to work. I'm blogging about the experience: Parts 1 and 2 are up now. follow along if you like! clockworkpenguin.substack.com/p/the-agenti...

Mucking about with #Suno again and it’s such a weird tool/platform/thing. Apart from turning the ‘normal’ slider up to ‘chaos mode’, it’s pretty impossible to glitch out. I tried nonsense style prompts and it hooves towards coherence. Must be a system default?

Tis the season for every Melburnian to become subsumed into their puffer jackets. Despite being one of said Melburnians, I genuinely don’t understand how you’re not all poaching in there. Also doesn’t the swish drive you bananas??

Sacre bleu ! C’est magnifique !

The problem with getting on top of your emails is that people will often respond to those emails and the inbox is somehow a ship of Theseus as well as a Sisyphean incline.

YOU’RE LISTENING TO SISYPHUS FM, THE HOME OF NON-STOP ROCK

There are many problems with genAI. But a great many of them come down to a general lack of understanding/literacy. Knowing what these systems are and what they can/not do goes a long way to dispelling much of the uncertainty and fear around them.

Herewith a piece from Meg Thomas and I in The Conversation, on the latest Oscars AI rules, and the long history of automation and ML in film production theconversation.com/most-mainstr...

Did I re-emerge from months of zero posts with a re-post of a joke that could be boiled down to “Se7ered Talking Heads”? You’re damn right I did.

[Se7en] DAVID BYRNE: [looks inside box] this is not my beautiful wife

My article “A media-materialist method for interpreting generative AI images” is now published online in Convergence. This piece represents the foundation for my genAI work over the last 2-3 years. Four layers: data, model, interface, prompt. journals.sagepub.com/doi/10.1177/...

Oh hello. I know I’ve been quiet, but it’s been a head down bum up kind of start to the year. Having said that, I’ve managed to set up a remotely accessible AI suite, vibe coded a family album browser, and install and trial a tiny AI agent on a ESP32 board. So a few blogs likely coming soon!

New post on generative systems and how working with rituals, recipes, and protocols has helped me feel into the ethics and values that influence and motivate this work. #genAI #ethics #creativity #authenticity #AILiteracy #technics #chance clockworkpenguin.substack.com/p/rolling-wi...

📝 New blog post: How I Read AI Images — in which I share a method for interpreting AI images based in the material mechanisms of their generation. Think Kittler, Parikka, and Cubitt. Full article coming soon, I hope! #genAI #AIimages #AIart #research clockworkpenguin.substack.com/p/how-i-read...

I am here for seagulls who sound like humans doing terrible impressions of seagulls

New Slop Stack post — A Little Slop Music: The AI experiment that turned my ick to 11 (now you can try it too!) Thoughts on creativity, authenticity, contested histories, paratexts, storytelling, and all that jazz. No, really; there's jazz in there too. open.substack.com/pub/clockwor...

I haven’t seen a Bad Guys. 1 or 2. But there’s a lil person in it with a tiny laptop. Who’s making tiny laptops for lil tech guys? Does Apple have a special LIL Division? Lilnovo?

I was interviewed by Bogdana Rakova about my generative AI research—why I embrace glitch, mess + failure as tools for learning. From origami-code mashups to the Slopocene, we explore playful, critical ways of rethinking creativity + AI literacy. open.substack.com/pub/speculat...

“Friction can happen not only when things break, it can also be intentional to prevent things from breaking. What if we could make the safe things easier (less friction) and the high-risk things harder (more friction)?” Check out this new series by Bogdana Rakova! 👀

A recent conference presentation worked to formalise some of my experimental work with genAI, and early film theory was the key to unlock some of those doors. I also draw on ancient philosophy, cybernetics, co-intelligence, and mixed-fidelity ritual practices. Enjoy! youtu.be/QDpaV1h0VQc?...

Does it have to be thinking OR genAI? Literature OR genAI? Other tech OR genAI? Ethics OR genAI? The planet OR genAI? Education OR genAI?

Brisbane in a couple of weeks: I’m speaking at the Digital Temporalities symposium at UQ, alongside visiting the @qutdmrc.bsky.social and the indomitable genAI Lab at QUT—this time with @clockworkpenguin1.bsky.social. Let’s see what kind of thinking the city allows.

After four years of work, our book Confronting the Climate Crisis is out now! It brings together activism, media, and ecoaesthetics to rethink how we engage with environmental collapse. Huge thanks to my co-editor Rebecca Najdowski and all our amazing contributors! link.springer.com/book/10.1007...

I wrote about one of my 'weird AI' experiments, 'spectral linguistics'. I prompt LLMs to 'remember' forgotten languages. They may hallucinate vocab, syntax, symbols! We learn a lot about how LLMs attune not just to language, but human sense- and meaning-making. clockworkpenguin.net/2025/07/28/s...

Yes. Yes yes yes yes yes. @ncotemunoz.bsky.social on the money again. This is what most of those who can’t understand why I — a critical thinker and lover of language, artists’ rights, and creative expression — use AI. And why and how it’s genuinely changed my life.

I wrote a thing for @theconversation.com on embracing and making slop as a means of developing critical genAI literacy and having some fun along the way. Enter the Slopocene! theconversation.com/understandin...

post slop ergo propter slop

I and my AI co-pilot have finally cracked. Simultaneously.

Didn’t sleep well lastnight. Felt rubbish and heavy and sinus infecty. But somehow today I got some stuff done. I worked closely with my little AI ecosystem, designing ritual systems, getting a little lost in lore and conceptual tangents.

After ten years of failed pitches, I finally got a commission for The Conversation. Exciting! But writing my first draft means I probably have enough material for my next journal article lmao

Apparently folks think that writing that contains an em dash is a dead giveaway that it's from ChatGPT, which I guess is a sign that most people have never read anything in their lives.

Expressions of Interest: Join us for the Re/Framing Field Lab in Brisbane/online (July 4): a hands-on co-lab exploring genAI, critical media practice & creative research. Discuss, debate, co-create: pitch, partner & prototype at the messy intersections of theory, practice & speculative design.

🧠 New blog post: Why can’t you just THINK?! Thinking isn’t the problem. It’s sorting. Starting. Sustaining. Three low-stakes ways to try genAI as a thinking partner — plus ethical offline model recs. 🔗 clockworkpenguin.net/2025/05/20/w...

💠 New blog post: On Clearframe—the anti-productivity productivity method I accidentally on-purpose co-hallucinated/created with generative AI clockworkpenguin.net/2025/05/14/c...

Come on, AI. What I want is very simple.

Here’s some of the ritual AI weirdness I’m playing with at the moment — more to follow soon. vt.tiktok.com/ZSh8uaywG/

When do we stop whispering/screaming about #genAI tools, and start including them in the larger category of ‘technology’? When do we sew up the hole this fun/scary new thing poked into our blanket, and accept it as part of the broader fabric of lived experience? clockworkpenguin.net/2025/04/29/a...

By way of update: I'm presently playing with #genAl through #glitch and ritual, trying to open up weirdness by leaning into hallucination as a feature rather than a bug. I'm also just about to try and build a posse of #tinyAl bots that do pointless things. Lmk if you’re paddling in similar pools!

ChatGPT generated a sticker to celebrate me finishing marking the first assignment for one subject. So that's a pretty neat encapsulation of my relationship with genAI.

Some of the posturing on social media from academics around the LibGen search thing is hilarious. "How could this happen?! This is something I've never seen before! Of course I don't use it myself and never have! I couldn't also name three similar sites nor would I ever bookmark them!"

⛓️💥 New blog post: How I broke Claude. Messing around with sophisticated language models; prompting nonrepresentational outputs; creative technical engagements with AI to generate alternative practices and narratives. clockworkpenguin.net/2025/03/18/h...

Can’t stop here.

All the responses I’ve read or seen to the Amazon book downloads policy change show a huuuuuge gap in digital literacy and awareness and understanding of DRM and the mechanisms of platform capitalism. Also — this is what happens when you click “agree” unthinkingly on T&Cs.

The amount of 'verify you're a human' tests I had to go through just now to renew my ChatGPT Plus subscription is the epitome of irony.