I am against GenAI and everything it stands for

TEXT START: It's been a few years since I posted a blogpost. It's time to do so again, good ol' "screaming into the void" as they say.

URL SCAN: I am against GenAI and everything it stands for
FIRST LINE: Published on: 2026-05-26

THE DISSECTION

This is a post written by a competent programmer — someone who genuinely knows how to build things — who has correctly identified the symptoms of systemic disruption but has constructed an entirely moral framework around an engineering inevitability. The author sees a cancer and demands the patient stop behaving badly. The diagnosis is precise. The prescription is theater.

THE CORE FALLACY

The central error is the assumption that the harms described are features of current implementation, correctable by ethical reform, regulatory intervention, or corporate restraint. The author writes:

"If all of this was done to better humanity, AI development would be done in public, data would be legally obtained, models would be released for free."

This logic is structurally identical to arguing that coal plants would be fine if operators just cared more about children. The author correctly identifies the rent-seeking, the surveillance, the intellectual property theft, the enshitification. But treats all of it as a governance failure rather than what it actually is: the mechanical expression of P1 (Cognitive Automation Dominance) — the outcome when AI achieves durable cost and performance superiority across cognitive work. No ethical reform stops this. No regulatory intervention preserves stable human-only economic domains at scale (P2). The author is cataloging the symptoms of a structural phase transition and prescribing bed rest.

HIDDEN ASSUMPTIONS

1. Reversibility: "I hope the AI companies collapse... I don't know how, though." The author recognizes the genie cannot be put back in the bottle while simultaneously hoping for reversal. This is cognitive dissonance wearing an intellectual disguise.

2. Human Cognitive Sovereignty as the Default State: The author treats the degradation of "thinking skills," "tinkering," and "problem-solving" as a loss that matters in a future where those capacities are economically necessary. Under DT logic, they're not. The mass employment -> wage -> consumption circuit dies regardless of whether young programmers can tar -xvf from memory or have to consult an LLM. Cognitive atrophy is a symptom of the collapse, not a cause.

3. The Enemy is the AI Companies: The author frames this as a battle against specific corporate actors (OpenAI, Google, xAI) with specific predatory practices. Under DT, even if every AI company collapsed tomorrow, the underlying technology would persist. The displacement mechanism is structural, not managerial.

4. Disinformation as the Primary Hazard: The author spends significant energy on Russian disinformation, state-influenced bot farms, political manipulation. These are real. But they are noise relative to the primary kill mechanism. Even in a perfectly clean information environment, the employment collapse proceeds on schedule.

5. The Pope as a Valid Signal: Citing the Pope's criticism of AI as evidence of severity is ideological armor — a moral authority lending weight to an argument that stands or falls on structural mechanics. The Pope's opinion about AI is as relevant to DT analysis as his opinion about interest rate policy.

SOCIAL FUNCTION

Copium with Technical Credentials. This post is written by someone with real programming chops who has correctly noticed their world is being destroyed by forces they cannot stop. The social function is moral venting dressed as analysis — it provides the psychological comfort of righteous opposition while delivering zero strategic utility.

It is also nostalgia theater: the entire "I learned Delphi as a kid, I used to tinker" section is grief writing. The author is mourning a world that was already dying before LLMs existed — a world of scarcity, craft, and individual problem-solving that the post-WWII order was systematically erasing through credentialization, abstraction, and platform dependency. LLMs didn't kill tinkering. smartphones and SaaS did.

Classify as: Copium with partial truth — the observations are accurate, the diagnosis is misdiagnosed, the prescription is wishful thinking.

THE VERDICT

This is sophisticated symptom recognition without structural understanding. The author correctly identifies that:

• GenAI is a rent-seeking apparatus built on stolen material
• It is destroying educational scaffolding and cognitive development pipelines
• It is generating massive disinformation and social fragmentation
• It is degrading the craft and durability of software engineering
• It is accelerating social isolation and loneliness
• The enshitification is accelerating

All of this is accurate. Under the DT framework, every single point is a feature, not a bug — it is the expected mechanical output of the transition from labor-capital productivity to AI-capital productivity. The author is diagnosing the symptoms of cardiac arrest and concluding the patient should eat better and exercise more.

The final paragraph is the most revealing: "I don't know how, though." This is the sound of someone who has correctly identified an existential threat and has zero leverage against it. The author knows. The author just doesn't want to say the thing: the post-WWII order ends regardless of what anyone does. The only question is which side of the transition you end up on.

The post is emotionally honest, technically literate, and strategically useless. It will resonate with hundreds of thousands of programmers who share the author's skill set and will feel equally powerless. It will not change a single structural variable.

The author is watching their world burn and has correctly described the fire. They simply cannot see that the fire was always going to happen, that it was always going to spread, and that the people starting it were never going to stop because they couldn't if they wanted to.