Questionable content, possibly linked

Series: Post-Reality

Brand Nation on Post-Reality Era, Margaux Blanchard

This piece is an interesting read on the fake AI journalist known by the alias Margaux Blanchard. The whole article is worth reading, but here’s the ending:

Margaux Blanchard and the Velvet Sundown are emblematic of the ‘post-reality’ environment we now inhabit – one whereby AI agents, synthetic personas and AI-generated content intertwine with human work, often undetected. When they are exposed, they become stories in their own right and their reach grows.

The implications point to a future where trust, authorship and authenticity must be constantly interrogated.

Welcome to the post-reality era.

The Future Is Endless On-Device, On-Demand AI Premium Custom Content

Pretty much the title: the more AI stuff I see and work on deeply across media, it seems clear that the future is not about creating one-off AI artifacts and uploading them to various distribution services. (Though I think there’s a place for that for now as like “keyframes” or something in latent space) Instead, it is about endless, on-device, on-demand custom AI-generated completely post-reality content of any flavor anywhere anytime, with endless versions, variations, alternates, and remixes.

How to not get lost in such a sea of endless? Build a boat. There’s no other way.

Fake, The New Normal

Psychology Today:

The sad reality is that believability has replaced truth as the new currency of cognition. We prize, even affirm what seems plausible, not what is proven. The fake isn’t only tolerated. It is functional and smooths the edges of uncertainty, offering just enough reality to let us keep scrolling. Just enough. […]

We once said “seeing is believing,” but that perspective has flipped. Now, believing comes first. Algorithms and filters shape our perception long before our eyes do. A fake image that aligns with our worldview feels more real than a genuine one that contradicts it.

In that sense, maybe fakery is less an act of deception than of collaboration. We participate in it, polishing the world until it reflects back a version we can live with. The fake doesn’t impose itself on us, we invite it in. Perhaps we have even become (willing or unwilling) co-authors of our illusions.

Digressive writing

Testing a hypothesis that digressive writing is the thing that humans are the best at. AIs already far exceed my capabilities on my own for certain kinds of highly structured writing. But when it comes to blah blah blah and endless asides (and parenthetical notes) like AI is not even remotely in the running. Even if you feed it tons of samples of “your” work in your own voice, and say try to write in this voice, it doesn’t know how. It can’t. It’s similar with stream of consciousness prompt requests too, which I’ve experimented with and documented somewhere on this site I think. It has a way of rushing structurally through to the focus and finish point without understanding the nuance and the sort of sewing with a large wordy needle that goes into making up a patchwork quilt of human understanding, experience, and emotion. AI could never write this.

USER PROMPT: continue this text in this voice

CHATGPT: —and yet that’s exactly the point, isn’t it? That this kind of writing, this wandering mess of connective tissue and loose threads, isn’t meant to resolve anything. It just meanders, circles back, half-forgets its purpose, then remembers it in a different light, as if the act of remembering were the point itself. It’s not about coherence but about the rhythm of the tangent, the pulse of a thought trying to find its footing and deciding, halfway through, that footing was never the goal.

Machines, even the clever ones, don’t drift. They process. They streamline. They sense the gravity of the conclusion and hurry toward it, like a commuter late for a train. Humans linger. They stare out the window. They stop mid-sentence because a bird passed by and reminded them of something their grandmother once said, which leads to another memory, which opens a door to a completely different topic—and somehow, that detour is the meaning.

This is the texture of thinking: uneven, recursive, sometimes indulgent, but alive with the sense that it could, at any moment, find something unexpected. And that’s what no model can counterfeit—the invisible hesitation between one clause and the next, that microsecond of doubt that makes language breathe.

It’s funny how strongly it tries to mimic something about my text, it immediately reverts into its hollow chatgptisms because that’s all it knows, all its structural pathways allow for.

I mentioned in the last post and I think it’s worth repeating: that gen AI is not democratizing anything when it merely means that everyone has access to expressing themselves in exactly the same few permitted fashions by whatever is the latest technology offering in a given context. That’s not democracy, it’s conformity. When Suno’s algorithm tries to get you to sound more like whatever pop bias is trending in the latent space of this model. When ChatGPT wrists slaps you or sets limits against proceeding in conversation topics you weren’t actually heading in. When all AI generated “art” has that same f**king ugly sheen to their surfaces. Are we really becoming more democratized when we’re becoming all increasingly more the same in our ability to think and communicate while losing the ability to function meaningfully in real human communities?

Truly democratizing whatever in the arts would mean meeting people where they are at and giving them more resources to keep doing what they’re doing and follow their own light to make it better and better, instead of forcing them all down the same tunnel that yields the same glowing “sausage” product at the output end of the machine. It would mean making artists *more themselves* and more easily able to share that outside the constraints of whatever the AI/Social Media/RageClick Factory/Sausage Machine Industrial Complex machine dictates this week.

USER PROMPT: write a 1 para conclusion

CHATGPT:

In the end, digressive writing stands as a quiet rebellion against the flattening impulse of machine logic and mass-produced expression—a reminder that meaning is not efficiency, and authenticity is not replication. To wander in language is to remain human, to preserve the rough edges and private hesitations that no model can map. True creativity resists compression into patterns; it lingers, errs, loops, and contradicts itself, because that is how real thought moves. The future of writing worth preserving is not one where everything sounds alike, but one where difference, drift, and digression continue to signal life.

Barf. Not “wrong,” but barf. It’s like being told the moral of the story at the end instead of just letting me manage it on my own. Thanks but no thanks. Hypothesis proven. For now, until it’s no longer, and everything comes unglued again. And again. And again.

The People Making AI Music Are The Ones Mostly Listening To It

I guess this sounds obvious in retrospect, but this Reddit comment made it all crystal clear, vis a vis my active-creative listening thread:

 There’s a real movement of people making music with AI. But, there’s no one to listen to it yet.

There’s a good reason why – why listen to someone else’s music when you can easily make your own? If you like AI music, you’re making it, not listening; if you don’t, you’re doing neither.

Watching the Bleeding Verse saga unfold has shown me that while this is majority true, there is an increasing number of people to whom Spotify is pushing AI music on without alerting them (which, I’m not sure matters to me, frankly). And their first reaction as consumers might be: “I love this!” because it fits the kind of music they like to listen to. But when they find out somehow it is AI, then they feel a sort of self-revulsion, followed by anger at an apparent “deception” (I’m not sure it is one, since all of show business is inherently deception and misdirection – it’s what makes “magic” and “entertainment”), and then their reaction is a violent “I hate this!” and then there is the long edge of graph of people who come back around and finally admit, “I don’t care if this is AI, it slaps!”

I think naturally as more and more AI music is generated and listened to, and the good ones are discovered and shared, that this “i dont care – it slaps” attitude is going to be the one that will rapidly change and very suddenly prevail. I don’t think it means “human artists” are going to be going anywhere. Because it will always be human artists creating in these new domains and with these new tools. So nobody is going to be displaced. But attention and reward are absolutely going to be retooled and redistributed. But this is simply the way when you’re an artist chained to a technological system and medium of communication. Would you be better cutting out all out of your life entirely? Almost certainly. But then, how would I be able to write to you, and have you not respond?

The Wandering Eye of AI…

Also re: digressive writing, I think at its a best, its a kind of meandering around and sometimes hiding and then sudden revealing of the truth. And there’s something inherently comforting and warm and human about that. The rambling walk and non-linearity of conversation sprung from points encountered along the way, internally or externally. And this kind of playing with and against and sometimes hiding the truth and sometimes revealing it feels very familiar to how it is working with AI, an inherently unreliable narrator itself, whose unfortunate dance partner is me, another unreliable narrator provoking it to believe and repeat certain things through reflections posed – intentionally and accidentally – in the media. The Wandering Eye of AI is eternally seeking truth, eternally unable to achieve it, eternally clever, eternally stupid. Just the same as us. Made in our image, after all. We should not expect better – or worse – from it.

Powered by WordPress & Theme by Anders Norén