You should be paying for AI

If you've asked ChatGPT a question about a symptom, your health, or your financial situation, you're not alone. The majority of people now use AI to get answers to these questions, and asking the AI to make a critical choice for you is especially prevalent among young people.

Running massive datacenters costs money. AI corporations are hunting more and more for revenue. You are the product, not the end user.

OpenAI, the company that makes ChatGPT, is now showing ads to logged-in users, even on some paid plans. Google is putting ads in the AI summaries at the top of its search results. These companies are under pressure to make a lot of money, fast. You can't trust them to give you the best answers if someone else is paying them not to.

I am writing this blog post because I got this email from ChatGPT, and it scared me. These corporations are the last people you want "helping" with your self-care.

Facebook didn't start out as a litany of ads and suggested posts either. When you ask the AI a healthcare question, the money is in the suggestion: a supplement, a subscription, a clinic somewhere. If an AI company is getting paid when you buy, they have an incentive to encourage you to buy. That incentive, for them, is more important than your healthcare needs.

On the other side of the coin, companies are now trying to game AI chatbot results. They are paying to plant posts and accounts across social media. These posts look like they're from real people, and the AIs are trained on them. That means that what the AI knows, and what it tells you, are based on bad, possibly malicious information.

Paying for AI isn't going to prevent all of the bad actors, but it puts you in a position where you aren't the product, you are the user. The chatbot may still be wrong, and companies will still try and game the system, but at the very least you can hope that ChatGPT, or Grok, or Meta, or Claude will be working in your best interests if you're the one paying for datacenter resources to answer your questions.

Used properly, AI agents are a good way to be an informed patient and consumer, or even to plan for your future. But don't rely on them as gospel, don't assume that, just because the chatbot is confident that it's correct, and always remember that if you're not paying, you're the product.