Read a brief explanation of what ChatGPT is before checking it out either directly online or by connecting with it in one of many software applications. Then learn the keys to mastering ChatGPT: perfecting the prompt and using the right ChatGPT plug-ins. Using ChatGPT will significantly increase your productivity, but you must always fact-check its responses before relying on them.
Don’t worry if you make a mistake along the way or find yourself stumped over what to do next. You're not alone. ChatGPT is completely new and mysterious for most people. Keep this cheat sheet handy and dive right in!
How ChatGPT works, in brief
ChatGPT is an incredibly advanced type of AI model, but it’s woefully short of the type of AI you see in sci-fi movies. It does not think; it predicts.
To embarrassingly oversimplify a major technological achievement, ChatGPT is like a really fancy predictive text tool. It analyzes your prompt and predicts what words will follow. In other words, you’re prompting it to complete a pattern you created with your prompt.
It might complete that pattern in exquisite and minute detail, matched wonderfully to context and intention and executed in perfect human mimicry. But just like predictive text, sometimes it predicts correctly and sometimes it does not.
You should always check ChatGPT’s work. And you must guard against any potential negative consequences that its use may cause. These consequences could be serious and should never be taken lightly.
Any prompt you enter and any conversation that results in ChatGPT is kept by OpenAI to train future AI models. That warning is clearly and prominently posted on the ChatGPT user interface. Your work here is your own, but it is not private.
Where to find ChatGPT and its competitors
You can access the free ChatGPT version or buy a subscription to the ChatGPT Plus premium model directly from OpenAI.
You can also find ChatGPT embedded in applications, many of which you likely use every day. Examples include Microsoft’s Bing, Teams, Office365, and Dynamics.
ChatGPT competitors include Bard, which is embedded in Google Search, and Quora’s Poe, which opens access to a range of chatbots, including ChatGPT, in the popular Q&A platform. You can even use Poe to build your own ChatGPT-like bot.
For advanced users, ChatGPT and the GPT models that drive it are all options you can use in a variety of ways. Look to the relevant APIs and plug-ins to help you connect the models with your applications.
You can also work with OpenAI and other sources (such as Databricks’ free Dolly 2.0 and Microsoft’s DeepSpeed Chat) to build models you can connect to your databases and train for your own purposes.
Master the ChatGPT prompt
The secret to success with ChatGPT is in mastering the prompt. Ask a question or state an instruction and set it off on a mission. Typing your question or command seems so simple, yet it’s the hardest part of using ChatGPT. Why? Because people tend to ask questions that are too general or vague or to give instructions that are too simple to get much more from ChatGPT than a regurgitation of a Wikipedia entry. Getting ChatGPT to deliver the real goods takes much more effort.
If you give it a basic prompt, you will get a bare-bones answer that you could have retrieved using a search engine such as Google or Bing, or a simpler AI assistant like Siri, Alexa, or Google Assistant.
This is why some people go no further with ChatGPT. They simply see no advantage in using it over other existing technologies. But the fault of that misjudgment falls squarely on the user, not AI.
Many people turn to ChatGPT to automate tasks and perform work processes, both of which it can do, but that’s too limited a view of its capabilities. If your aim is to optimize the use of ChatGPT, consider ways to use it to help you think better, harder, more accurately, and more creatively. View ChatGPT as augmenting you, rather than just a means to automating your work.
In short, you’re going to need to think more, not less, to use ChatGPT to its utmost best. The cheat in using ChatGPT is “think harder, work easier.”
Here’s a handy list of other tips to help get you started on the path to mastering the art of the prompt:
- Plan to spend more time than expected on crafting a prompt. No matter how many times you write prompts, the next one you write won’t be any easier to do. Don’t rush this part.
- Start by defining the goal. What exactly do you want ChatGPT to deliver? Craft your prompt to push ChatGPT towards that goal. If you know where you want to end up, you’ll be able to craft a prompt that will get you there.
- Throw out any limits for this tech that you’re harboring. What will hold you back are the imagined limits that you project on ChatGPT.
Think like a storyteller, not an inquisitor. Give GPT a character or a knowledge level from which it should shape its answer. For example, tell ChatGPT that it’s a chemist, an oncologist, a consultant, or any other job role.
You can also instruct it to answer as a famous person would, such as Churchill, Shakespeare, or Einstein, or as a fictional character, like Rocky. Give it a sample of your own writing and instruct ChatGPT to write its answer to your question or complete the task in the way that you would.
Remember that any task or thinking exercise (within reason and the law) is fair game and within ChatGPT’s general scope. For example, instruct ChatGPT to check your homework, your kids’ homework, or its own homework. Enter something such as computer code or a text passage in quotation marks and instruct ChatGPT to find errors in it or in the logic behind it.
Or, skip the homework checking and ask it to help you think instead. Ask it to finish a thought, an exercise, or a mathematical equation that has you stumped. The only limit to what you can ask is your own imagination and whatever few safety rules the AI trainer installed.
- Be specific. The more details you include in the prompt, the better. Basic prompts lead to basic responses. More specifically, concise prompts lead to more detailed responses, more nuanced responses, and better performance in ChatGPT’s responses.
Use prompt chains. This technique uses multiple prompts to guide ChatGPT through a more complex thought process. You can use multiple prompts as a single input, such as telling ChatGPT it’s a team consisting of several members with different roles, all of whom are to answer the one prompt you entered.
You can also use multiple prompts in a sequence in which the output of one becomes the input of the next. In this case, each response builds on the prompt you just entered and the prompts you entered earlier. This type of prompt chains form organically, unless you stop ChatGPT from considering earlier prompts in its responses by starting a new conversation.
- Use prompt libraries and tools to improve your prompting. Some examples follows:
- Check out the Awesome ChatGPT Prompts repository on GitHub.
- Use a prompt generator to ask ChatGPT to improve your prompt.
- Visit ChatGPT and Bing AI Prompts on GitHub.
- Use a tool such as Hugging Face’s ChatGPT Prompt Generator.
- Try specialized prompt templates, such as a curated list for sales and marketing use cases.
Tons of curated lists are in repositories on GitHub as well as lots of free prompting tools from a variety of sources. Just make sure that you double-check sources, apps, and browser extensions for malware before using or relying on them.
If you’re concerned about data privacy, and you should be, consider elevating your game by graduating from ChatGPT to training a GPT model of your own. Why? Two big reasons are to prevent giving proprietary data to the third-party that owns and trains ChatGPT or any other large language AI model and to dedicate your model to specific tasks necessary to your business or other endeavor.
ChatGPT has limited storage space, so unless you’ve adapted it with a plug-in or something similar, it will drop your earlier prompts to make room for new ones. If you want to keep these conversations, take screenshots or copy and paste the conversations to a document you can save on your device or in the cloud.
Use ChatGPT but don't trust it
Any generative AI model, ChatGPT included, can and does lie. AI trainers are working on a fix, but it’s damn tough to do considering the very thing that makes these generative AI models work so well is the same reason they fail so badly on occasion.
ChatGPT creates outputs. It can, therefore, also create lies, called hallucinations in AI-speak. These lies can seriously harm your job, your business, your community, your country, and possibly your species one day. But so too, theoretically at least, can the person sitting to either side of you at work or on a bus, train, or plane.
To compete in an increasingly fast world where production quotas and expectations are soaring, you’ll be at a distinct disadvantage if you don’t use AI like ChatGPT to assist you with your work. But if you use it without double-checking its outputs, you’re at considerable risk, too.
If ChatGPT succeeds in doing the work for you without your intervention, the model will likely replace you on the job soon. Or if ChatGPT makes a massive error, you’ll be at risk of losing your job for failing to manage its results.