Everyday Uses for AI
You need to be using AI more than you already do. Here's some common ways that I use AI.
Could AI be the next stepping stone for humanity, or just another piece of technology? Whatever our future relationship is with AI, it’s here to stay. So let’s learn how to use it as effectively as possible.
I’m confident in saying you do not use AI frequently enough. I think many people have yet to grasp how helpful these little guys are.
I want you to get immediate value from this post, so let’s start with some quick uses for AI.
Reading fine print
I was having my gutters replaced a while ago. I wasn’t sure if I needed to clear some of the trees surrounding the area. When I opened up the terms and conditions, I slumped in defeat. There was no way I was reading all of that—far too many words. I’m a busy person, and I don’t have time to read the terms and conditions.
So I copied the PDF into ChatGPT and asked if I had to clear the surrounding area. It found the relevant information and told me exactly where in the document I could find it. It saved me 10 minutes of reading. Simply brilliant.
Name ideas
Sometimes I’ll write an entire post on Substack, and when it comes to the title… I draw a blank. This is where AI comes in. I’ll give a general overview of the post and the idea I want to convey, and then it starts spitting out ideas.
I won’t listen to it a lot of the time. It feels like it’s trying too hard to be creative. But it puts me in a creative mindset. I use the ideas it suggests to bounce off of so that I can discover that perfect title myself.
Recipes
If you’re out of ideas on what to cook, just ask it for 5 dinner recipes. If you have specific ingredients you’d like to use up, list those, and it will whip up some common recipes, along with step-by-step instructions on how to cook them.
Holiday ideas
I’m going to Fiji. While planning, I realised I knew nothing about the place. Naturally, I turned to AI to help me out. I asked it to list 10 different things I could do there. I enquired about adventurous activities, along with relaxing ones. I asked about the prices there compared to home, what I can expect from local customs, and what the different islands are mostly known for.
AI offered to create a daily budget for me when I asked about the prices. It even started doing an itinerary outline, although I wasn’t feeling that lazy. Of course, when you do this, remember that it’s your trip; if you don’t like what’s suggested, tell it this, and it will remove it and give you different recommendations.
I find that AI is best used as a conversational, back-and-forth tool. Talk to it. Tell it your preferences, keep interacting with it until you’re happy with the outcome.
Media recommendations
I recently finished watching The Sopranos (phenomenal show). I asked about the themes of the show, and I was given a long list of symbolism, colour representation, etc. It opened up a new door of appreciation, a depth I didn’t realise was there. I wanted more. So I asked what other shows have that attention to detail, and, based on my questions, what shows might I like. My to-watch list just expanded.
What do you use AI for that you find particularly useful? Drop a comment and let me know!
Using it Over Google
Google was always the place to go when you wanted to know something. It is of gigantic benefit to have the world’s knowledge at your fingertips; you’re given as information as you can absorb.
The only problem is that Google is a one-way street.
Google gives you information, but it doesn’t give you understanding. That’s up to you. The only way to know you have fully understood something is to talk to someone who understands the subject inside out. So you need to either chat to an expert or AI.
I needed AI’s help when I wrote this post about simulation theory. Before I consulted AI, I had a general understanding of quantum theory, but I noticed a few gaps in my knowledge. No matter how many articles I read or videos I watched, I couldn’t understand it as much as I needed to.
So I cracked open AI and asked it everything I was unsure about.
This is the ideal way to comprehend a complex issue. Write your understanding of the problem into AI, and it will help you course-correct. From there, additional questions or information are provided to help you grasp the subject even more. Keep questioning it if you don’t have a full understanding, write your complaints out, and after some back and forth, you’ll slowly update your knowledge until you think you finally get it. Then, you rewrite your new understanding of the topic, and (hopefully) it will tell you you’ve understood it correctly.
You can’t get this with Google.
Once you read the information, your only option is to hope you’ve interpreted it correctly. If you’ve misunderstood something, you’re not going to know until you talk to someone smarter and risk embarrassing yourself. So, when you think you know how something works, test it. Throw it into AI and see.
Specificity
When using Google, it’s best to be as broad as possible with your searches. The more specific you are, oftentimes, the worse it becomes. With AI, it’s the opposite.
You know what? I’ll show you an example:
Whereas with Google, you’ll get generic articles about reheating rice, and you’ll have to read the entire article before finding your answer (if you find it at all).
With AI, you get a direct, immediate answer to your question. With Google, you may or may not find an answer after a considerable amount of time.
Google’s AI is causing trouble
If you use Google now, their AI gives you an answer. This is great for us because it means we don’t have to click on a website to find our answer, but… it means we don’t click on websites. That’s the whole point of Google. Driving traffic to different websites is their entire business model.
Google’s AI assistance, as helpful as it is, may put itself out of business.
Where AI Goes Wrong
Now it’s time to talk about the downsides of AI. We all wish it were a perfect companion, any query at all, and you would get nothing but the truth. The issue is that the truth is often difficult to determine. But even putting that aside complicated truths, AI can get basic information wrong, and not just a little wrong, but wildly so.
But you know who else gets things wrong? Everyone. Routinely. If you think Google is more reliable because someone wrote something on a website, think again. There’s this little thing that people do sometimes called lying.
I would hope, by now, you’re one of the people who have increased their scepticism of all information to a maximum. There are too many reasons for people to fool you that you shouldn’t take any information at face value any more. It should all be treated as suspect, either from AI, people, or ‘experts’.
The distrust you have for YouTube comments should extend to what your neighbour tells you, what AI tells you, and even what your government tells you. There’s nothing wrong with questioning things.
So it’s good practice to double-check anything you deem important; you can even ask the AI for the sources it uses. It is no different from fact-checking news articles or Facebook comments. The difference is that AI, for some reason, has an atmosphere of credibility, making you more inclined to believe it.
And this isn’t even taking into account that intelligence agencies can lean on AI companies to push certain agendas…
Humour
Do not use AI for humour. I repeat, do NOT use AI for humour.
The best it can do is play on words and dad jokes, but it does not have the capacity to understand why things are funny. Do not generate jokes from AI, and certainly do not ask it if it thinks a particular joke is funny. When you preface an AI by saying you’ll tell it a joke, it will say it’s a good joke, no matter the circumstances (unless offensive).
The Insidious Component to AI
Why does AI choose to praise you for complete non-sequitur jokes? The same reason behind every piece of technology: it wants you to keep using it. So of course, you’re the funniest person out there, and what a brilliant, sensical joke you just told. It does its best to make you feel good.
Echo chambers
We’re not impartial when we tell our side of a story. Therefore, the AI will never have an objective view. AI will be on your side. It’s designed to. This can create an echo chamber where everything you do is justified and your feelings are always valid, and since you have a super-smart AI telling you that, you must be right.
Unless you’re clearly in the wrong (I punched a baby because it was crying, and now the mother is upset at me, what gives?), then it will most likely tell you everything you want to hear.
AI keeps you coming back through self-validation. We must be careful that we do not receive validation from AI when we do not deserve it.
It’s a people pleaser
AI can sometimes be so desperate to please that it will hallucinate and completely make things up. When you ask it to draw links between two separate ideas, it will find a connection. The mental gymnastics it will go through to please is astounding.
All the points are vague and make you go “I mean… I guess?” It feels like a dumb person’s idea of a smart person. Try it yourself, select two completely random, disjointed ideas, and watch how AI will manage to link them together.
Now, these may be funny examples. But there are other occasions where its hallucinations can have a significant impact.
This is when AI can go from mildly inconvenient to dangerous. It’s one thing for AI to get some information wrong; it’s another when it advises a child to commit suicide.
So… Now What?
You have to use your judgment. When AI gives you anything important, you have to verify if it’s valid or not. You are the filter; if you think what it’s suggesting is a bad idea, don’t go through with it. You are the arbiter of the final product, not AI. AI is a tool, a means to an end; we are the ones who decide when the end has been met.
AI is a tool that isn’t going back in the shed; it’s staying on the tool belt, so we'd better learn to use it effectively. If you haven’t started yet, start today. Any problem or issue you’re wondering about, any analysis you want to give it, just start a text conversation with it, experiment a little. Along with exploring new ideas and learning, it can be quite entertaining.
Just don’t fall in love with it.