ideas, startups, ai,

Use ChatGPT to Pitch Almost Any Idea

Adam Dunkels Adam Dunkels Follow May 02, 2023 · 6 mins read
Use ChatGPT to Pitch Almost Any Idea
Share this

In a previous article we looked at a simple way to pitch almost any* idea. Today we’re looking an equally simple way to use ChatGPT to automate the process.

* I have no data to support this hypothesis

The mechanism we use in the previous article the following simple yet effective pattern for presenting ideas:

  1. Describe the problem
  2. Describe the consequences of the problem
  3. Describe the solution
  4. Describe the consequences of the solution

Despite this pattern being very simple, it can still be tricky to use it to express ideas.

This is where AI can help.

Modern AI is all about generating things

Large Language Models (LLMs) are a type of generative Artificial Intelligence (AI) models that have recently become extremely popular because of the ChatGPT chatbot.

The way these AI models work is that they generate text based on a user-supplied prompt. The prompt is a piece of text that guides the model as to how it will generate its text. How to best construct these prompts is an art form in itself, dubbed prompt engineering.

Let’s teach ChatGPT our pattern

ChatGPT and other similar systems are good at learning a pattern and then generating text based around that pattern. And what we have is a simple four-sentence pattern for presenting ideas. So let’s try and see what ChatGPT makes of it.

Let’s provide this prompt to ChatGPT:

Procosoco is a technique to describe ideas in which the idea is expressed as four sentences.
The first sentence states the problem.
The second sentence describes the consequences of the problem.
The third sentence states the solution.
The fourth sentence describes the consequences of the solution.

Express the idea "let's go to lunch" using the Procosoco technique.

And this is what ChatGPT produces:

Problem: We are currently feeling hungry and need to eat.
Consequences of the problem: If we do not eat, we may become fatigued or unable to focus on our work.
Solution: Let’s go to lunch.
Consequences of the solution: We will be able to satisfy our hunger and feel energized, which may increase our productivity and ability to focus on our tasks.

This is remarkably similar to the example that I produced for the previous article:

  1. We are hungry.
  2. This makes us grumpy and unproductive.
  3. Let’s go to lunch.
  4. When we’re back, we’ll be happy and productive again.

So let’s try it with the second example I provided, a fictional AI model that could run on a slide rule. This is the prompt:

Good. Now describe the idea "SlideRuleGPT", which is a revolutionary GPT that is able to be computed entirely on a slide rule, using the Procosoco technique.

This is the output:

Problem: Current language models such as GPT-3 require advanced hardware and computing power to operate effectively, which can be expensive and not readily available to everyone.
Consequences of the problem: This can limit access to these powerful tools for those who cannot afford or do not have access to such technology.
Solution: SlideRuleGPT is a revolutionary language model that can be computed entirely on a slide rule, which is a low-cost and widely available mechanical computing tool.
Consequences of the solution: This breakthrough in language modeling technology may democratize access to advanced computing tools, allowing more people to engage in natural language processing tasks and accelerate innovation in the field of AI.

Again, this is strikingly similar to our human-generated example from the previous article:

  1. Today’s AI models are extremely expensive to train and run.
  2. This makes generative AI available only to a few large corporations.
  3. We present SlideRuleGPT, a new AI model that can be run on a single slide rule
  4. This makes generative AI available to anyone, even without Internet access.

So ChatGPT is able to both understand out the pattern and to figure out what a matching problem is to our proposed solutions.

Let’s try to apply this to some other random ideas that we might want to convince people, or ourselves, about:

Idea: “Use version management instead of emailing Word files”

Problem: Currently, we're emailing Microsoft Word files back and forth to collaborate on documents.

Consequences of the problem: This can result in confusion, with multiple versions of the same document being edited simultaneously, leading to inconsistencies and errors. It can also be time-consuming and inefficient, with a lack of clarity over who has made what changes and when.

Solution: We should use a version management system, which allows us to work collaboratively on documents in a controlled, organized way. This will enable us to keep track of changes and ensure everyone is working on the same version of the document.

Consequences of the solution: This will lead to greater clarity and efficiency in document collaboration, with a reduced risk of errors and inconsistencies. It will also enable us to easily see who has made what changes, and when, streamlining our workflow and improving productivity.

Idea: “I should sleep more to feel better”

Problem: I'm not getting enough sleep.

Consequences of the problem: This can lead to a range of negative consequences, such as feeling tired and sluggish during the day, difficulty concentrating, and a weakened immune system.

Solution: I should prioritize getting more sleep, by establishing a regular sleep schedule and ensuring that I get at least 7-8 hours of sleep per night.

Consequences of the solution: This will lead to a range of benefits, such as feeling more alert and energetic during the day, improved cognitive function, and a stronger immune system. It may also help to reduce the risk of developing chronic health conditions such as heart disease and diabetes.

Idea: “Use a password manager”

Problem: Keeping track of multiple passwords for various accounts can be difficult and time-consuming.

Consequences of the problem: This can lead to the use of weak, easily guessable passwords, or the use of the same password for multiple accounts, which can increase the risk of a security breach.

Solution: Using a password manager can help to securely store and organize passwords for various accounts, making it easier to use strong, unique passwords for each account.

Consequences of the solution: This can improve overall online security and reduce the risk of a security breach. It can also save time by eliminating the need to remember or write down multiple passwords, and by providing a centralized location to manage passwords.

Conclusions

ChatGPT turns out to be a remarkably powerful tool to express ideas using the simple technique from the previous article. By supplying a simple prompt that teaches ChatGPT our general pattern, we can simply ask it to explain a simple idea with this pattern. And the results are pretty good!

Adam Dunkels
Written by Adam Dunkels Follow
I am Adam Dunkels, IoT consultant. I build exceptional IoT products together with my customers. I am known for being the original author of the lwIP stack, the guy behind the Contiki OS, and the creator of protothreads, among other things.

Want my help to build your next connected IoT product? Get in touch!