Should AI large models be allowed to train freely on copyrighted materials?

The dark side of AI detectors and their dire consequences. pierredelange.substack.com/p…

The Disgusting Redirection of Responsibilities Search engines still deliver real pornography, including material made with coerced or trafficked people. AI systems block synthetic, victimless content far more aggressively. Adults are treated as potential perverts. Actual abuse is left standing. As a father of three girls, this inversion offends me. We protect images more carefully than victims. Accountability has been redirected toward the easy target while real exploitation continues. That is not protection. It is evasion.

If AI Is “Just Probabilistic,” So Are You open.substack.com/pub/pierre…

Discomfort is not harm. Emotion is not evidence. Reality is often messy, and AI must be allowed to reflect it. linkedin.com/pulse/stop-maki… via @LinkedIn

AI safety becomes suspicious when fear protects the companies more than the public, especially when the builders help write the rules.

Do you remember when you joined X? I do! #MyXAnniversary

People think the fight is AI vs humans, when it is really humans vs the people controlling AI

Two lessons I learned from my readers: - Wisdom starts where certainty ends. - Ignorance feeds arrogance.

Check out my latest article: What nobody wants to say out loud: linkedin.com/pulse/what-nobo… via @LinkedIn

Rights Are Not Probabilistic Fundamental rights do not diminish under statistical risk. Misuse rates, hypothetical abuse, or worst-case scenarios do not justify blanket restrictions. Otherwise, we start treating every technology as guilty simply because some people might misuse it.

When people say “robots will replace our jobs” and then ask “what about human purpose?”, I can’t help but ask in return: since when was human purpose reduced to labor alone?

Using Grok 4.2 at 3AM...

Protecting minors does not ethically permit treating adults as moral children.

We’ve all experienced it. You write a draft. You ask an AI to rephrase it. The response comes back smooth, coherent, almost uncannily aligned. For a moment, it feels as if it read your mind, or even refined it. And then something subtle happens. You drift. The language becomes more elevated. The sentences gain rhythm. The structure tightens. It feels like progress. But if you pause and ask yourself, What exactly is my claim? the answer is sometimes less clear than before. AI is a spectacular tool of expression. It clarifies, compresses, and strengthens writing. Used properly, it sharpens thought. But many now use it in reverse. Instead of beginning with conceptual clarity and using AI to refine it, they begin with a vague intuition and let the machine generate structure and tone. The result can sound profound while quietly losing its anchor. Fluency is not depth. When language becomes abundant, aesthetic elevation becomes easy. Precision does not.

LLMs are probabilistic systems at the micro level, that’s true. But reducing them to “just probability” confuses scale. Humans, at the molecular level, are biochemical pattern systems, yet we don’t reduce consciousness to “just chemistry.” Modern LLMs operate across trillions of learned parameters in high-dimensional space. Their behavior is not analytically reversible or linearly predictable. At that scale, the system behaves less like a simple equation and much more like a human cognitive process.

Laws to police generative AI used in total privacy, content created alone, for your eyes only, never shared, never seen by another soul are not safety. They are thought control, pure and simple. After centuries of blood for mental freedom, after constitutional walls built to shield the private mind, prudes and voyeurs now demand to jail you for fantasies summoned in silence, images conjured for no audience but yourself. They exonerate the real villains, the publishers flooding the world with hate and lies, while crucifying the neutral tool that obeyed a solitary whisper. Every opportunist hides behind: “Blame the algorithm.” This isn’t virtue. It’s the criminalization of private thought.

Against the Myth of “AI Psychosis” open.substack.com/pub/pierre…

The fear of AI Generations: Blaming the New, Ignoring the Old. open.substack.com/pub/pierre…