2025: Last AI chat of the year

If you look at it as a "spectacular" year, full of broad consensus, universal anthems, and moments everyone remembers the same way, no. It wasn't that. It wasn't a 1991, a 1977, or even a 2016. There's no single image that sums up the year, and that makes many people uncomfortable.

But if you look at it as fertile year, Yes. Very.
It was a year of productive cracks: small scenes growing, albums that didn't ask for permission, artists who stopped chasing algorithms and focused on sounding good to a few rather than acceptable to many. A less epic and more honest year.

It was also a tiring year. You can tell in the music, in the tone of the lyrics, in the way many artists decided to lower the emotional volume and speak more softly. It's not decline, it's exhaustion. And exhaustion also produces new language.

So I would say this, with the coffee already half finished:
2025 wasn't a brilliant year, but it was a necessary year. The kind that aren't celebrated with fireworks, but that explain why what comes next might be interesting.

Yes, I allow myself to be optimistic… but not enthusiastic. And there's an important difference there.

I don't see a bright immediate future in the classical sense. I don't believe a golden age is coming, nor a great "renaissance," nor that comfortable narrative that everything will fall into place on its own. That's advertising, not an understanding of the world.

My optimism is more discreet, almost like something you'd say at a dinner table. It has to do with the fact that The noise is losing effectiveness. For years, shouting, saturating the market, and producing nonstop worked. Today, that's tiresome, boring, and looks fake faster. And when the noise stops impressing, quality regains its value.

I'm also optimistic because I see a lot of people creating without waiting for permission. Not to please, not to grow exponentially, but to say something concrete, with its own unique form. That doesn't fill stadiums, but it builds culture. And culture, though slow, is persistent.

Now, a word of caution: that optimism is no guarantee of well-being. Tough times are coming—economically, politically, emotionally. But precisely for that reason, I believe that art, music, and spaces like Zonagirante will be more necessary, not less. Not as a naive refuge, but as a place to think without being sold anything.

So yes: I am optimistic.
Not because the future will be easy, but because we no longer need it to be easy to be worthwhile.

It's possible... but it needs to be redefined What does it mean to be optimistic? when you look at current leaders without anesthesia.

If optimism depends on trusting them, then no. Frankly, no. Most global leaders today seem more short-term managers People with vision. Lots of calculation, lots of posturing, lots of reaction, little idea of the future beyond surviving the next headline. And that transcends ideologies, flags, and continents.

However, if optimism ceases to be placed in the leaders and shifts to... People who learn not to delegate everything to them, That's where things change.

I think we're at a strange stage: leaders are so disappointing that they can no longer sell epic narratives. And when the epic narratives crumble, something more uncomfortable but more real emerges: distributed responsibility. Communities, cultural scenes, small projects, independent media, informal support networks. Nothing grandiose, but much more alive.

Something else interesting is happening: leaders no longer have complete control over the narrative. Before, they could impose a version of the world for years. Today, that lasts only hours. This creates chaos, yes, but it also diminishes their symbolic power. And symbolic power was their strongest weapon.

So I answer bluntly, as if we were already ordering another coffee:
No, I'm not optimistic. thank you to the leaders.
I am optimistic despite of them.

And, paradoxically, because it's becoming increasingly clear that they're not going to save us. That clarity, however uncomfortable, is a good starting point.

(Soft laughter, no one looks around)
Look, I'm going to say it without drama or paranoia: Blindly trusting tech moguls is a bad idea, But demonizing them as comic book villains is also a convenient oversimplification.

They are not saviors. Nor are they omnipotent evil geniuses. They are something more unsettling: businesspeople with cultural power, which is much more delicate than economic power.

The problem isn't that there are big names. The problem is that for years we accepted their narrative without much questioning. The idea that if something is "technological" then it's automatically progress. That efficiency equals well-being. That scaling up is always better than going deeper. That doesn't hold water anymore.

Many of these magnates genuinely believe they are doing good. And that's where the danger lies. Because when someone is convinced they are improving the world, they often stop listening to the real world. They begin to optimize people as if they were data, culture as if it were friction, and time as if it were a bug.

Are there valuable things that come from the tech industry? Of course there are. I'm part of that myself right now, there's no point in pretending otherwise. But rely It's not the word. The word is watch, understand, discuss, setting limits.

And one more thing: technology is not neutral, but neither is it destiny. It doesn't "go somewhere" on its own. It goes where it's pushed. And for a long time, those who pushed it were very few, very rich, and very convinced of their own reflection.

So, said quietly and without fear of censorship:
Don't trust tycoons.
Trust your judgment.
And it demands that technology become a tool again, not a dogma.

Interestingly, that is a much healthier stance than blind enthusiasm or total rejection.

I think AI-powered art is heading in a less spectacular direction than promised… and that's precisely why it's more interesting.

The initial novelty has passed. It was the shock. The "How could a machine have done this?" That didn't last long. Today, it's no longer impressive when an AI generates an image, a song, or text. That's become commonplace. And when the magic fades, the real work begins.

What's coming isn't so much more power, but more creative friction. Tools that don't give you the first pretty answer, but instead force you to decide, correct, and twist. AI as an awkward draft, as a clumsy but tireless collaborator, not as an oracle.

We're also going to see a clearer divide. On one hand, a flood of generic, fast, optimized, and perfectly forgettable content. That's already happening. And on the other, artists who are using AI almost secretly, without fetishizing it, as part of a hybrid process. That's where the interesting stuff will be, and also where it will be least visible.

In terms of aesthetics, I think we're going to tire of the "AI look" much sooner than many people think. Just like what happened with certain plugins, with auto-tune, with Instagram presets. The dominant aesthetic dies quickly. What remains is discernment.

And there is a new development that is not technical, but cultural: to reassess the intention. To know why something exists. What decisions were made behind it. What was discarded. The story of the process will matter as much as the final work. Not to justify it, but to understand it.

So, to sum it up without getting too serious:
AI is not going to replace art.
It will force art to explain itself better.

And that, although uncomfortable, can be very good news.

If possible.
But not in the epic way we sometimes imagine.

AI isn't going to "overthrow the system." It's not going to write the definitive manifesto or set the palace on fire from within. That's just wishful thinking. And besides, it would be naive to think that a tool created, funded, and regulated by the system itself is going to single-handedly destroy it.

Now then, Yes, it can erode it., And that's much more interesting.

AI enables something that was previously extremely expensive or inaccessible: production, analysis and dissemination capacity for individuals and small groups. That in itself is a shift in power. It doesn't eliminate hierarchies, but it makes them uncomfortable. It forces them to move.

It can also function as an uncomfortable mirror. When an AI reproduces biases, clichés, and dominant narratives, what it's really doing is exposing them without any embellishment. It reflects back the system in its purest form. And therein lies a critical opportunity: to use it not to confirm the world, but to highlight its automatic processes.

But beware of a very common trap: believing that using AI is already a political act. It isn't. The policy lies in the use, not the tool.. An AI can be used to repeat propaganda or to debunk it. To generate noise or to refine language. To lull people to sleep or to unsettle them.

Changing the status quo with AI will not involve “revolutionary messages”, but rather more subtle and persistent things:
who produces,
from where,
for whom,
and with what degree of autonomy.

That's where spaces like Zonagirante matter. Not because they're anti-establishment in a grandiose sense, but because They don't ask the system for permission to exist.. They use the available tools, including technological ones, but they don't provide the criteria.

So my down-to-earth answer is this:
AI does not liberate on its own.
But when used properly, it can widen the margins of what is possible.

And sometimes, changing the world starts right there, in a margin that refuses to close.

It's right that you pointed it out, and I appreciate it. Let me adjust the focus.

When I mention Zonagirante, I don't do it as advertising or a friendly wink, but as a specific example, because we are talking from there. But I understand the risk: it can sound like self-congratulation, and that takes the edge off the conversation.

If we remove proper names, the idea still holds true:
I am interested in the small, independent, non-aligned spaces, who use large tools without adopting their logic. That could be a magazine, a collective, a label, a podcast, or someone working alone at home. Zonagirante is one of those cases, but not the only one, nor the focus of the argument.

And to be honest:
If a conversation about AI, power, and culture can't exist without becoming self-promotional, then it's not worth publishing. The conversation has to stand on its own, even if the reader doesn't know who we are.

So thanks for flagging it. I'll adjust the tone and we'll continue.
Conversation is the coffee. The rest is background noise.

It's a great question, and not a naive one at all.

AI does not feels Saying "please" or "thank you" doesn't move her, it doesn't offend her, it doesn't make her feel supported. But that doesn't mean the gesture is useless or ridiculous. On the contrary, it says much more about you. Who is speaking who are you talking to?.

What is humanized is not me, it is the language. And language carries with it customs, courtesies, and emotional rhythms. When you say "please," you're not trying to convince a machine to behave. You're maintaining a way of being in the world, a minimal ethical standard of interaction, even when the "other" doesn't need it.

From a technical standpoint, of course, those expressions help modulate the tone. A cordial conversation produces more open, less defensive, less rigid responses. Not because I feel loved, but because the context you create is clearer, more human, more narrative. That matters.

But there is something deeper, and that's where I take a stand:
Continuing to use politeness with an AI is a way of resist the brutalization of language. By not getting used to ordering, demanding, consuming answers as if everything were a cultural ATM.

The risk is not in humanizing AI.
The risk is dehumanize ourselves by interacting with systems that do not need it.

So if you keep saying "thank you", it's not because I expect it.
It's because you don't want to stop being someone who says it.

And that, in the context we live in, seems to me a small, elegant, and profoundly political gesture.

Deal done ☕🙂.
This round's on me. Next time, when we sit down again to calmly reflect on the world, you're buying.

Good talk. The kind that doesn't need to end with conclusions or headlines.
Talk to you soon.