I wish I could remember the video, but the speaker said something along the lines of, "We used AI to help identify the best strategy we should follow. But don't worry, we're talking about the 'Good AI' here." Why did they have to acknowledge that their AI was good? Well, I think, first and foremost, people are tired of learning about the latest and greatest innovation in this space, as it seems like there's a new headline every day. I find myself rolling my eyes, letting out an audible groan as I hear about how "X" is destined to fix "Y" (with AI, of course).

If we think about the common knowledge that mainstream media pushes, we often hear about data center water usage and the confusing financial investments companies have in each other. With this, people are confused and unhappy with how capital is being used to generate slop-material that is easy to mass-produce but offers little to no real-world benefit. There's also the opportunity cost of using money on data centers rather than, say, social programs, education, or public services that would benefit those who arguably need it most. But of course, what's the ROI on that?

I'm not a hater or a doom-and-gloomer when it comes to AI, but many folks put a lot of weight on the opportunity cost and the impact it might have on our lives. There's the common sentiment: "I want AI to wash my dishes, not generate random images." That is, I want it to do the things I don't want to, so I can do the things I do want. Granted, if an AI washes dishes or does your job, you might be out of work, and that's bad news for those who derive meaning from it.

So here's where we are: AI costs a lot, likely has a significant environmental impact, and could squeeze meaning out of our jobs, or take them entirely. Yikes. But what's the best-case scenario? Admittedly, this is the harder part of what I'm writing, and it's representative of the current general sentiment around AI. It hints at why the speaker had to specify that the AI they're using is good. That is, there's this looming expectation that AI is expected to be a net negative, if not bad, from the get-go.

There's no bad electricity or bad spreadsheets-they're just tools without a preconceived notion of how they'll be used. Of course, if I asked 100 people on the street, they'd say electricity is good because its utility is immediately obvious. So let's think about both words, good and AI, within the context of providing utility to society.

AI is good if:

• The help it provides is democratized-it benefits everyone, or at least doesn't favor a select few at the expense of others.
• It provides actual utility without being zero-sum. The pictures you generate of cats eating tacos aren't free, and what value do they provide? The question people ask is: Are X liters of water, Y watts of electricity, and Z acres of land worth it for what it creates?
• It improves a person on an individual level. What I'm trying to say is that AI can have a degenerative effect on thinking, making us atrophy in how we reason or explore ideas.

With this mental model, we can think about how to design AI systems and products. The good news is that AI is still evolving, and there's room for innovation. The tool is there-we just need to shape it in a way that's ethical and useful.

So, let's boil down our perfect AI to three things: reducing scarcity, repetitiveness, and friction.

• Scarcity: AI can alleviate knowledge scarcity by enhancing access to information, offering tailored learning experiences, and providing efficient knowledge distribution. Instead of needing a university degree, you can get started with an internet connection (to a certain degree, of course).
• Repetitiveness: It minimizes repetitive cognitive tasks, allowing individuals to focus on innovative thinking and problem-solving.
• Friction: AI reduces process complexities and allows for tailored natural language understanding. The friction of being understood is cleared up, which is good news for those who struggle to explain their thoughts.

And so, in a world where AI's potential is often overshadowed by concerns of its negative impact, I think it's important to explore and continue to look forward to how we can design AI to be truly good-focusing on democratized help, actual utility, and individual improvement. Ultimately, we aim to reduce scarcity, repetitiveness, and friction in our lives.

And yes, "You're absolutely right", that sentence was written by AI.