A week using AI for everything
Decisions, writing, shopping and work: what got better, what got worse and where I stopped thinking.
The main disadvantages of artificial intelligence are that it gets things wrong with total composure, that it doesn't understand your situation unless you…
The main disadvantages of artificial intelligence are that it gets things wrong with total composure, that it doesn't understand your situation unless you explain it, that it can compromise your privacy, that integrating it properly costs money and time, and that it produces generic content if you don't guide it. None of them makes it useless, but all of them are worth bearing in mind before trusting it with something important.
A good part of these disadvantages comes from how it works: AI doesn't "know" the way a person knows; it generates the answer that seems most likely. That's why it can give two different answers to the same question, not know what happened after its training or make up a quote. And there are disadvantages beyond your screen: energy consumption, the effect on some jobs and the risk of learning less. What usually confuses people are generic lists of pros and cons: what matters is your case. A serious mistake in a legal report is irrelevant when you're asking it for name ideas.
Lists of disadvantages are fine, but I learn more when I trip over them. To see them in practice I did the challenge "A week using artificial intelligence for everything", consulting it on decisions, texts, purchases and work. There I tell you what got better, what got worse and which things I stopped asking it for and which I still do.
Decisions, writing, shopping and work: what got better, what got worse and where I stopped thinking.
One email a week with what was verified, what was refuted and what is still on the bench. No news roundups.