AI is transforming our world at lightning speed, and tools like ChatGPT are now part of everyday life. While we’re adapting to these innovations, AI agents are already making their debut. Staying ahead means experimenting, learning, and embracing new opportunities. Beyond adapting, experimenting with AI unlocks chances to empower yourself and expand your skills.
My journey with AI
Two years ago, I began exploring AI’s potential when consumer applications first emerged. My journey started with fun side projects, like creating images with the first edition of DALL-E—an experiment that made me a finalist in The New Vermeer, a TV art competition. I also used AI to produce podcasts, including a fascinating discussion between two fictional Americans about my Dutch book on human-centred transformation.
The real value: building custom GPTs
The most significant breakthrough came when I started building GPTs. These custom AI tools solve specific problems and save time on routine tasks.
My first GPT? A tool that creates customised packing lists for different destinations based on planned activities and travel seasons. This simple yet practical solution saves time, and it hints at the future possibilities: AI agents planning and booking entire trips, leaving us only to review and approve.
Last month, I built a GPT to SEO-proof my web copy. This tool has been a game-changer, saving me both time and money with every new blog or portfolio update.
How do you start building a GPT?
Building a GPT is empowering. Why? Because it allows you to create a tool tailored to your specific needs. The best part? You don’t need coding expertise, just curiosity and a willingness to experiment.
Here’s a step-by-step guide:
1: Define your purpose
- Identify a recurring task or challenge, such as automating a packing list or improving SEO content. Clearly articulate your goal.
2: Access ChatGPT
- Use OpenAI’s ChatGPT platform, ideally the Plus version with GPT-4 for better performance.
3: Draft initial instructions
- Write a clear prompt detailing what the GPT should do. Include examples and rules to guide its responses.
4: Test and refine
- Run your GPT with sample inputs. Analyse the output and tweak the instructions to improve results.
5: Iterate and optimise
- Experiment by adding or revising functionalities. Test repeatedly until the outcomes are consistent and reliable.
6: Save and share
- Save your GPT for regular use. Share it with others if it could help them too.
Step 7: Learn and expand
- Keep exploring new features and build more advanced tools as your skills grow.
Building your own GPT isn’t just practical; it’s empowering. Let your curiosity guide you, and you’ll soon discover how to refine your prompts and enhance the AI’s performance. Each step builds your skills and confidence, giving you an edge in a tech-driven world.
Not getting the hang of it? Get in touch and let’s see if I can help.
