AI for the Reluctant: What Are ChatGPT, Claude, Perplexity, and Gemini Actually Good For?

The Best AI Tools for Beginners: ChatGPT, Claude, Perplexity & Gemini

If you’ve managed to avoid AI so far, I can’t really blame you. Depending on what you read or watch, AI is either going to revolutionize everything, take away everyone’s job, destroy the internet, make it impossible to know what’s real, or somehow usher in the end of civilization.

Then, in the middle of all that, someone tells you that you really need to start using ChatGPT.

It’s a lot to take in.

The thing is, you don’t have to decide whether AI is going to save the world or destroy it. You don’t even have to be particularly enthusiastic about it. There are some legitimate concerns about AI, including privacy, misinformation, copyright, job displacement, and the fact that these systems can sometimes produce remarkably convincing nonsense. There’s another side to the story that gets lost in all the hype.

AI can be incredibly useful and you don’t need to become an AI expert to take advantage of it.

So what is AI actually good for?

The easiest way I’ve found to think about generative AI is as a very capable assistant who happens to communicate through conversation. You can ask it to explain something, help you write something, research a topic, organize information, brainstorm ideas, analyze a document, plan a trip, troubleshoot a problem, or simply help you think through something you’re stuck on.

That last one may actually be one of its most useful capabilities.

For years, we’ve been trained to think about computers in terms of commands. You open an application, find the right menu, learn the right terminology, and then figure out how to make the software do what you want.

Artificial Intelligence works differently. You can tell it what you’re trying to accomplish, even if you don’t know exactly how to accomplish it.

You can say, “I don’t understand this. Can you explain it to me?” You can follow that with, “OK, but what does that actually mean for me?” Then you can ask another question. And another.

You don’t have to know the right question before you start.

That’s a surprisingly big deal.

Which AI should you use?

There are dozens of AI tools now, but four names come up more than any others: ChatGPT, Claude, Perplexity, and Gemini.

The confusing part is that they can all do many of the same things. They can write, answer questions, analyze information, research topics, and help you brainstorm. The differences aren’t as dramatic as some of the marketing would have you believe.

Still, each has developed some particular strengths, and understanding those differences can help you decide where to start.

ChatGPT is probably the best place to start

ChatGPT is the one most people have heard of, and there’s a good reason for that. It’s become an unusually broad, general-purpose AI assistant. You can use ChatGPT to write and edit, brainstorm, research, analyze spreadsheets and documents, look at images, create images, work through technical problems, write code, plan trips, learn new subjects, and a whole lot more. It can also search the web when you need current information.

One of the things I like most about ChatGPT is how easy it is to use for things that don’t fit neatly into a category.

• Suppose you have a problem at work but aren’t quite sure how to describe it. You can explain the situation in your own words and ask what you might be missing. Or maybe you’ve received a complicated letter from an insurance company. You can give it to ChatGPT and ask it to explain what it means without all the jargon.
• Suppose you want to plan a trip and want a full itinerary. You can ask for the most popular places or the places off the beaten path. You can then ask it to export the entire itinerary to a spreadsheet.
• Suppose you want to plan a dinner party for a special event and want ten passed courses along with wine or cocktail pairings.
• Suppose you want to cook dinner and only have a piece of frozen fish, half a tomato, ginger, garlic, and soy sauce. ChatGPT can give you 5 or 500 reliable recipes.

You can also upload your own files and have it work with them. If you have a spreadsheet you’re struggling to understand, a PDF that’s 40 pages longer than it needs to be, or a collection of notes that you’ve never managed to organize, that’s where AI starts becoming more than a novelty.

If you don’t know which AI tool to try first, I’d start with ChatGPT simply because it can handle such a wide range of tasks.

Claude is particularly good with writing, complicated information, and coding

Claude is made by Anthropic and is probably ChatGPT’s closest competitor.

Again, there’s a tremendous amount of overlap. Claude can do many of the same things ChatGPT can do. But I’ve found that Claude is particularly appealing when the task involves a lot of reading, writing, editing, analysis, or coding. However, Claude is not good at creating images. If you need to do that, I recommend Gemini.

For example, if you have a long document and want help making sense of it, Claude is very good at working through large amounts of information. It’s also a strong writing partner. You can give it something you’ve written and ask it to make the argument clearer, identify weak spots, or rewrite it without completely stripping away your own voice.

Claude has another feature called Artifacts that makes it interesting for people who want to create things rather than simply have conversations. It can produce documents, diagrams, web pages, visualizations, and working prototypes in a separate workspace. That last one is particularly interesting. You can describe something you’d like to build and Claude can sometimes create a working version of it for you, even if you’re not a programmer. That’s a pretty significant change in what it means to “use a computer.”

If your work involves a lot of writing, documents, analysis, or coding, I’d definitely give Claude a try.

Perplexity (Microsoft) is the one I’d use when I need to research something

Perplexity is a little different from the other three.

The simplest way to describe it is as an AI-powered search engine. You ask a question in normal language, Perplexity searches the web, and it puts together an answer based on what it finds. Most importantly, it shows you the sources it used.

That makes it particularly useful when you’re trying to learn about something you don’t know much about.

Let’s say you’re planning a trip to Brazil. You could ask Perplexity about the best neighborhoods to stay in Rio, the current requirements for American travelers, restaurants in a particular neighborhood, or the best way to get from one city to another. You can then follow the links and check the underlying information yourself.

Or maybe you’re considering buying a new laptop. Instead of searching Google for “best laptop 2026” and opening 14 tabs, you could ask Perplexity to compare several models based on battery life, repairability, performance, and reviews from reputable publications.

That’s where Perplexity really shines.

It’s also important to understand that Perplexity isn’t magically immune to the problems that affect other AI systems. It can still misunderstand a source or draw an incorrect conclusion. The fact that it provides citations makes it easier to check its work, but you should still check it.

If your first thought when you have a question is, “I need to research this,” Perplexity is worth trying.

Gemini makes the most sense if you already live in Google’s world

Gemini is Google’s AI assistant. Like the others, it can write, answer questions, research, analyze information, and help with all sorts of everyday tasks.

Where Gemini gets particularly interesting is its connection to Google’s ecosystem.

If your life revolves around Gmail, Google Drive, Google Docs, Google Sheets, and Google Calendar, Gemini can potentially work with information that’s already sitting inside those services. That’s a very different proposition from simply opening a chatbot and asking it a question.

Imagine being able to ask an AI to help you make sense of a collection of emails, summarize information you’ve stored in Google Drive, or work with a spreadsheet in Google Sheets.

For someone who already does most of their work inside Google’s ecosystem, that integration may matter more than whether Gemini is marginally better or worse than another AI at answering a general question.

So if you’re a heavy Google user OR you want to create images or illustrations, I’d give Gemini a serious look.

Which AI tool should you actually use?

If you’re looking for one AI tool that can handle a little bit of almost everything, I’d start with ChatGPT.

If you’re primarily interested in writing, editing, analyzing complicated documents, or building things, I’d try Claude.

If you’re researching a trip, a purchase, a current event, or just trying to get up to speed on a subject and want to see the sources behind the answers, I’d try Perplexity.

And if you’re already deeply invested in Gmail, Google Drive, Docs, and Sheets, Gemini may fit most naturally into the way you already work.

But there’s no reason you have to choose only one.

I use different tools for different purposes, and there’s nothing particularly strange about that. You probably don’t use the same piece of software for every task on your computer. AI is no different. I use the free version of ChatGPT as my general personal assistant, the $20 per month Claude as my general business assistant (often Claude is taking the place of an entire team of assistants), and Gemini as an image creator. I don’t use Perplexity very often but know people who do because they are tied to a Microsoft environment.

But what about all the bad things you’ve heard about AI?

Some of those concerns are completely legitimate. AI can be wrong. Worse, it can be wrong in a way that sounds incredibly confident. That’s probably the most important thing for a new AI user to understand.

If you ask ChatGPT, Claude, Perplexity, or Gemini a question and it gives you a beautifully written answer, that doesn’t mean the answer is necessarily true. These systems are very good at generating plausible language. That is not the same thing as having perfect knowledge of the world.

So don’t use AI as an unquestionable authority.

If you’re making an important medical, legal, or financial decision, AI can help you understand the subject and prepare questions, but it shouldn’t be the final word. Likewise, if you’re using AI to research something important, check the sources. If you’re using it to write an email that could cause a problem if it’s wrong, read the email before you send it. That doesn’t make AI useless. It just means you need to use some judgment.

In fact, I think the most useful way to think about AI is as a collaborator rather than an authority. It’s very good at helping you get started. It’s very good at giving you ideas. It’s very good at looking at a problem from another angle. It can save you an enormous amount of time. But you’re still the one making the decision.

You don’t need to learn how to “prompt”

There’s another thing that can make AI seem more complicated than it really is: prompt engineering.

You’ll find plenty of articles explaining the precise formula for writing the perfect AI prompt, complete with elaborate instructions about roles, context, constraints, and output formats. There are situations where that stuff is useful. But you don’t need to learn any of it to get started.

Just talk to it.

If you’re trying to plan a vacation, tell it where you’re going and what you like to do. If you’re trying to write an email, give it the email and explain what you want to say. If you don’t understand something, tell it that you don’t understand. And if it gives you a bad answer, tell it.

• “That’s too complicated. Explain it more simply.”
• “That’s not quite what I’m asking.”
• “Give me a few alternatives.”
• “What am I missing?”

That’s the part that makes AI different from a traditional search engine. You can have a conversation.

Start with something you actually need to do

If you’re skeptical about AI, I wouldn’t recommend starting with one of those lists of 100 amazing things AI can do. Most of them aren’t particularly relevant to your life.

Instead, wait until you have a problem.

• Have an annoying email you need to write? Give it to ChatGPT and ask for help.
• Have a 30-page document you don’t want to read? Ask Claude to summarize it and tell you what deserves your attention.
• Planning a trip? Ask Perplexity to research it.
• Have a bunch of information scattered across Gmail and Google Drive? Try Gemini.
• Or ask AI something you’ve always wondered about.

You might be surprised how useful a simple question can be.

One of the best things about AI is that you don’t have to worry about asking a stupid question. There is no person on the other end of the conversation rolling their eyes because you don’t understand something. You can ask the same question five different ways until it finally makes sense. And then you can ask another question.

AI doesn’t have to be amazing to be useful

There’s a tendency to talk about artificial intelligence in extremes. It’s either going to revolutionize everything or destroy everything. It’s either going to replace all of our jobs or turn out to be nothing more than glorified autocomplete.

The reality is probably going to be considerably less dramatic. AI is a tool. And like most tools, its usefulness depends on what you do with it. You don’t have to believe that AI is going to save the world. You don’t have to believe that it’s going to destroy it, either. You can simply decide whether it’s useful to you. Maybe you’ll use it every day. Maybe you’ll use it once a month. Maybe you’ll try it and decide that you don’t like it.

But if you’ve avoided AI because you’ve only heard about the bad stuff, I’d suggest trying it for something small and practical. Ask it to explain something you’ve never understood. Give it a recipe and tell it what’s in your refrigerator. Paste in an annoying email and ask it to help you respond. Ask it to help you plan your next vacation.

You don’t need to become an AI expert. You don’t even have to like AI.

You just might find that having a remarkably knowledgeable—and occasionally very wrong—assistant available whenever you need one is actually pretty damn useful.