With GPT Builder you can create your own personalised ChatGPT

In the fast-evolving landscape of artificial intelligence, generative AI tools like ChatGPT have emerged as frontrunners, captivating the imagination of technophiles and business leaders alike.

The allure of these AI marvels lies in their remarkable ability to simulate human-like interactions, offering insights, generating ideas, and even composing literary pieces. This growing demand reflects a broader trend in technology: the quest for personalised digital experiences that resonate with individual needs and aspirations.

Now, generative AI tools have taken a leap up to the next level with GPT Builder, a groundbreaking development in the realm of generative AI from ChatGPT creator, OpenAI, released last week.

GPT Builder not only lets you imagine what creating your own version of ChatGPT would look like, but also do it tailored to your specific interests, industry and personal styles of communication. GPT Builder represents a paradigm shift from one-size-fits-all AI solutions to a more personalised approach.

With this innovative creator, anyone from a small business owner, to a creative writer, to a PR practitioner, can fine-tune the capabilities of ChatGPT, forging a digital companion that not only understands their specific context but also augments their daily tasks with a touch of AI-driven ingenuity.

I’ll say it more simply: It’s possible to make a GPT just by chatting with ChatGPT and describing what you want!

It marks the beginning of a new era in AI customisation, where personalisation is not just a feature but the very core of the user experience.

How does GPT Builder work?

To start with, you’ll need the right version of ChatGPT, either ChatGPT4 Plus or ChatGPT Enterprise. Both are paid subscription versions – the builder does not work with the free version of ChatGPT. I use ChatGPT 4 Plus.

The GPT Builder is an interactive platform that lets you create and customise your own GPT model, ie, a personalised chatbot. This bespoke approach means that you can tailor responses to suit your specific industry needs, company ethos, or communication style. The Builder has a user-friendly interface, complete with drag-and-drop functionality that simplifies the complex process of AI programming with just a few clicks and adjustments. It shields you completely from any code.

Anyone can easily build their own GPT—no coding is required. You can make them for yourself, just for your company’s internal use, or for everyone. Creating one is as easy as starting a conversation, giving it instructions and extra knowledge, and picking what it can do, like searching the web, making images or analyzing data. Try it out at chat.openai.com/create.
OpenAI, November 6, 2023

In ChatGPT4 itself, you’ll see a new choice in the menu sidebar called ‘Explore’ that, when you click it, gives you the option to create a GPT of your own.

From this point, the Builder leads you through a series of questions where your responses guide the tool in creating the customised version of ChatGPT that will then be a choice in the sidebar menu in your ‘My GPTs’ list.

I went through the process of creating a personalised ChatGPT4. I told the Builder that my need was for a chatbot that would research potential topics to give me ideas for articles for my blog taking into account the topics focus of the blog and the chatbot personalisation I wanted. I called it ‘Idea Spark’, a name suggested by the Builder in response to the prompts I’d entered early on during the creation process.

This is the crucial part of personalising your ChatGPT that GPT Builder will create. Your instructions (answers to the prompts) are what will make your chatbot personal to you, in what it will do. Here’s a screenshot of some of the prompt/answer processes that led to the creation of ‘Idea Spark’.

Getting to this conclusive step took just fifteen minutes, and I then had a personalised version of ChatGPT4. I could leave it as this and start using it straight away, which I have done. But as this is a learning process, I will spend time tweaking ‘Idea Spark’ based on my experiences with it during the next few weeks.

Once you have your personalised GPT model, look into the ‘Configure’ section where you can make simple changes or add some content. Here is what ‘Idea Spark’ looks like today – I have made no changes yet; this is as it came after GPT Builder created it.

You’ll note in the screenshot above, at the top, it says “Published / Anyone with a link”. This means that Idea Spark is publicly accessible if you know the link. I’ve not yet made that open and likely will do so after my initial testing. But if you’d like to see it before that and give it a spin yourself (and let me know what you think of it), drop me a note on LinkedIn, Pebble or Threads.

Is a personalised ChatGPT4 better than the standard ChatGPT4?

That’s a question I had at the start, but I quickly realised that it’s not the right question to ask.

Given that your personalised version is a version of ChatGPT4, it’s technically equal. Where it differs is in its focus – what it allows you to do and what it pays attention to after you provide prompts and answers to its responses, and how it engages with you the user embodied by its tone and personality.

I see it this way:

For general use, not narrowly focused or zeroed in on a specific industry, product range, business, etc, where you want the widest focus to produce the most wide-ranging responses, I’d use the standard ChatGPT4.
For specific use, where you are focused on your industry or profession, and on topics of very specific interest to you and relevance to your industry or profession, I’d use a personalised version of ChatGPT4.

Therefore, one is not better than the other. Rather, each is better at the task you seek it to complete based on the wide or narrow focus you desire.

I think that time will tell if this thinking is correct or not.

GPT Builder: Communication transformer

Front of mind for me as I experiment with personalisation is how much of a game changer this will be in organisations.

First of all, the GPT Builder encourages communicators to experiment with different styles and formats, pushing the boundaries of conventional corporate communication. By offering an environment to trial innovative ideas and create personalities, it paves the way for a more adventurous approach to messaging, one that could help differentiate a brand in a crowded marketplace.

Moreover, the Builder’s capacity to blend data analytics with content creation enables the surfacing of insights that might otherwise remain buried in the depths of big data. This synthesis of quantitative and qualitative analysis can unveil new perspectives and opportunities for storytelling, helping many people across organisations narrate their data in a manner that is both enlightening and compelling.

This is equally valid in both internal and external communications. Imagine a personalised GPT for employee communication. Another one for investor relations. One more for public relations. Each of them specifically tailored for communicating and engaging with those distinct groups (let’s not just call them ‘audiences’).

If you’ve read this far, I imagine it’s quite clear to you that personalisation is at the heart of the GPT Builder’s appeal.

Standard, non-personalised, ChatGPT operates on a general model, proficient across a wide range of topics and styles. It lacks a nuanced understanding of specific industry terminologies or company-specific lingo. How could it be other than that, given its positioning as a generative AI tool that can be used by literally anyone with an Internet connection, asking it anything?

A personalised GPT created by GPT Builder can be trained on specialised datasets, encompassing industry-specific knowledge, company policies, and even the subtle nuances of a brand’s tone and voice. This enables the creation of a model that not only speaks the language of the business but also embodies its ethos and values.

Then there’s the adaptability of personalised models – truly a game-changer.

You can continuously update and refine these personalised GPTs to keep pace with the evolving dynamics of a business environment. For instance, as a company expands into new markets or adopts new product lines, the personalised ChatGPT model can be updated to include this new information, ensuring that its responses remain relevant and insightful.

This stands in stark contrast to the static nature of the standard ChatGPT Plus, for instance – while periodically updated by OpenAI, it cannot be personalised at an organisational level.

The GPT Builder gives you greater control over the AI’s response mechanisms, too. This is crucial for maintaining compliance with regulatory standards and ethical guidelines, especially in industries like finance, healthcare, and law. Organisations can tailor their personalised models to avoid specific topics, adhere to legal constraints, or align with cultural sensitivities, something that the one-size-fits-all approach of standard ChatGPT cannot do.

There are risks, of course. Data privacy and security, to name two. And there are ethical considerations, such as the potential for AI to perpetuate biases or misunderstand cultural nuances. Such important issues must be an equal part of the evaluation procedures in organisations.

All of this really just touches the surface of the capabilities and possibilities that a personalised ChatGPT can offer. More is coming from OpenAI in the months ahead such as a GPT Store from where you will be able to use personalised models created by verified builders. Read OpenAI’s Introducing GPTs for a good overview on how the company sees the immediate future.

I see opportunities for organisations where a GPT Builder-created GPT can offer a level of specificity, adaptability, and control that is unmatched. It empowers organisations to leverage AI in a way that is finely tuned to their specific context, paving the way for more effective, efficient, and personalised communication strategies.

It’s definitely worth kicking the tyres.

Additional reading:

Personalized A.I. Agents Are Here. Is the World Ready for Them? (New York Times, November 11, 2023)
Status report: artificial intelligence (AI) in pu b lic relations (Stephen Waddington, November 11, 2023)
Custom GPTs: New Ways To Utilize ChatGPT For SEO & Marketing (Search Engine Journal, November 10, 2023)
AI is about to completely change how you use computers (Bill Gates, November 9, 2023)

Related reading:

ChatGPT, Claude, Bing, or Bard? (August 29, 2023)
AI for global good (June 28, 2023)
It looks like PR has its head in the sand about AI (March 1, 2023)
As Bing AI search joins the chatbot space let’s look at some realism amongst the excitement (February 21, 2023)

(The generative AI high-definition photo at the top of the page of a woman sitting at a computer screen using GPT Builder was created with Adobe Firefly. I had also created photos using the DALLE-3 integration with ChatGPT4 that, while very good, showed a robot in the screen display every time even though I specifically asked in the prompt not to include a robot. This tech clearly makes up its own mind at times.)