I Have a PhD in Computer Science. Here’s Why I Hate AI

I spent the better part of a decade in a computer science PhD program. I’ve read the papers, run the training jobs, and sat in seminars where people far smarter than me argued about loss functions until midnight. And when someone at a dinner party says “I just hate AI,” I no longer assume they’re wrong. I assume they’re aiming at the wrong target — because so was I, for a while.

Is it really the science people are angry at?

No. Almost nobody railing against AI online is actually angry at gradient descent. They’re angry at what’s been built on top of it and shoved into their lives without asking. That distinction matters, because it’s the difference between a real conversation and a culture-war shouting match.

The same family of techniques driving the backlash also drives AlphaFold, which cracked a fifty-year-old problem in protein structure prediction, and machine-learning models now used in early cancer detection. Nobody is marching in the street against those. The public isn’t rejecting the mathematics — it’s rejecting the way specific companies have chosen to deploy it. Chatbots wedged into search bars you didn’t ask for. “AI features” that quietly harvest your documents. Customer service replaced by a hallucinating bot with no escalation path. That’s not computer science. That’s a product decision, made by executives optimizing for adoption charts, not for you.

I’ve come to think of this as the spine of the whole debate: the science versus the deployment. Conflate them and you either defend indefensible corporate behavior in the name of “innovation,” or you reject genuinely useful research because you’re furious at a chatbot that lied to you. Neither is accurate.

Why does this feel like a threat to survival, not just annoying?

Because it’s landing on top of an economy that already feels like it’s failing most people. Wages have been stagnant for years while housing costs have gone vertical, and that combination turns any large technological disruption into something closer to a threat than a novelty.

Compare this to the smartphone revolution. The iPhone launched in 2007, and the years that followed a broad economic recovery — homes were more attainable, labor markets eventually tightened, and disruption felt like it came with upside. AI is arriving in a completely different climate: precarious gig work, unaffordable rent, and a labor market where a single layoff can be catastrophic. When people say AI feels like it’s coming for their survival, not just their job title, that’s not hyperbole — it’s an accurate read of the economic conditions AI landed in. I wrote more about how this specific anxiety compounds in the piece on whether AI could run the economy — the fear isn’t really about intelligence, it’s about who controls the levers.

What are companies actually doing wrong here?

They’re forcing untested, unreliable AI into products people already depend on, without meaningful consent, and optimizing for adoption metrics instead of whether the feature actually helps anyone. That’s a deployment failure, and it’s where the real ethical rot lives.

Some of this is well documented and genuinely damning:

Layer onto that the sheer forcing of it: AI assistants defaulted on, autocomplete features that can’t be fully disabled, “smart” replies inserted into your email. None of that is you choosing to adopt a tool. It’s a company deciding your product experience will change whether you like it or not, because internal adoption metrics look better than user satisfaction scores. If you want the receipts on how badly this plays with the public, the Pew reality-check on Big Tech and AI is worth your time — trust in these companies is cratering for a reason.

Is AI inherently good, bad, or something else?

It’s neither. AI is an objective-driven tool — it optimizes for whatever it’s told to optimize for, and its value is entirely a function of the socio-economic system pointing it. That’s not a dodge; it’s the actual mechanism.

A model optimized to detect tumors early is a triumph. The same underlying architecture, optimized to maximize engagement or minimize headcount, becomes something people reasonably fear. The tool didn’t change. The objective function did, and someone chose that objective function. In a society with strong labor protections, universal healthcare, and a real safety net, automation reads as liberation — fewer hours spent on drudgery, more time for everything else. In a society where losing your job means losing your healthcare and possibly your housing within months, the exact same automation reads as an existential threat. Same technology, opposite emotional valence, because the surrounding system is different. This is the throughline connecting AI backlash to broader anti-AI sentiment — it’s not really a technology story, it’s a governance story wearing a technology costume.

1. Separate the math from the product

   Ask whether you’re angry at a technique (rare, and usually misplaced) or a shipped product decision (common, and usually justified).

2. Name the actual harm

   Scraped work, forced defaults, unaccountable hallucinations, and job displacement without a safety net are specific, nameable failures — not vibes.

3. Aim the demand at policy, not vibes

   Data rights, labor protections, retraining funding, and deployment regulation are achievable asks. “I hate AI” is not a policy position; it’s a shrug that changes nothing.

So what should “I hate AI” actually turn into?

It should turn into specific, structural demands, because venting doesn’t move a single regulator, and it doesn’t cost a single tech company anything. The lack of corporate accountability, the regulatory vacuum, and decades of poor economic policy are the real villains here — not the transformer architecture.

Concretely, that means pushing for: enforceable data rights over what trains these models on your behalf; labor protections and funded retraining for roles AI displaces; and deployment regulation that requires consent, disclosure, and liability when an AI system gets something catastrophically wrong. None of that requires anyone to stop researching AI. It requires the companies profiting from it to be accountable for how they ship it — which, notably, is exactly the accountability most of them have spent the last several years trying to avoid.

I still believe in the science. I ran the experiments, I’ve seen what the good version of this looks like. What I don’t believe in anymore is trusting a handful of corporations to deploy that science responsibly without someone forcing their hand. That’s not a contradiction. That’s the whole point.