Organizations are constantly seeking ways to enhance efficiency, from improving employee onboarding and training to speeding up customer support. Imagine a tool that integrates smoothly into these areas, using specific domain knowledge to create a seamless experience, all without the need for coding skills. The idea of a custom GPT model, perfectly suited to your organization’s needs, is not just a future possibility—it’s an accessible reality ready for deployment.

Reflecting on the past year reveals a remarkable evolution in AI’s role in reshaping business and creative processes. The introduction of advanced AI capabilities has not only revolutionized tasks ranging from research and idea generation to the creation of bespoke images for digital content, but has also redefined the boundaries of innovation and productivity. This AI progress extends beyond mere data summarization or content generation; it signifies a profound shift in how complex tasks are approached and executed.

“Anyone can build a customized chatbot by providing a handful of instructions and, in some cases, uploading a few documents”.

Further cementing this transformation is the latest update to ChatGPT Plus (note that this is the paid version, which costs $20 per month) that allows individuals, organizations or small business owners to seamlessly create personalized GPT-powered chatbots that can be customized for their industry, business or task. This breakthrough heralds a new era of AI accessibility, offering myriad possibilities for varied business scenarios and signaling a significant advance in the practical application of AI across industries. For example, as shared here in The New York Times, “the owner of a small bed-and-breakfast could build a chatbot that answers questions for anyone who stays there. You can imagine someone building a GPT that specializes in teaching someone a new language or gives interior design advice. Anyone can build a customized chatbot by providing a handful of instructions and, in some cases, uploading a few documents. When a bed-and-breakfast owner designs a bot that answers questions for guests, for instance, he or she might upload the B&B’s existing guest manual.”

In a separate New York Times article, Kevin Roose shared how he created a chatbot called Daycare Helper as way to navigate the specific rules for his child’s daycare. “They have this parent handbook that has all these rules in it. I’m constantly looking things up in the parent handbook.” In addition, as way to help him become smarter at investing, he was able to upload his grandfather’s investing book to create a custom GPT for investing and financial advice.

Tailoring AI to Personal Passions

After exploring several personalized GPT examples, I embarked on creating two of my own, driven by my passions for cooking, CrossFit and ice hockey. The first, a culinary-focused GPT named “X Minutes to Flavor”, accessible to those with ChatGPT Plus, was designed to aid in weekly menu planning. Traditionally, this Sunday morning task involved sifting through my ever-expanding cookbook collection and marking recipes with post-it notes. This GPT, tailored to Middle Eastern and Latino recipes from chefs featured in my collection, needed to adjust recipes for two instead of four. Its configuration, considering regional fare and preparation times of 30, 45, or 60 minutes, merely took a couple of minutes to create, followed by ongoing refinements.

“After requesting that “X Minutes to Flavor” generate a weekly menu for two, balancing proteins with vegetarian options and with each meal confined to a 30- to 45-minute preparation window, it promptly provided a comprehensive menu and a detailed shopping list organized by quantity and weight.”

After requesting that “X Minutes to Flavor” generate a weekly menu for two, balancing proteins with vegetarian options and with each meal confined to a 30- to 45-minute preparation window, it promptly provided a comprehensive menu and a detailed shopping list organized by quantity and weight.

My second creation, “CrossFit Hockey Coach,” stemmed from my quest to enhance my cardio and endurance for CrossFit and ice hockey. This GPT, based on a structured and tiered CrossFit program, curates workouts that build progressively and takes into consideration individual performance metrics, like completion times. It emphasizes safety, tailors workouts to varying fitness levels and avoids overly complex or risky exercises. The workouts are balanced with adequate recovery periods and are specifically designed to improve hockey skills. The GPT maintains an instructional and motivational tone, offering clear guidance and encouragement. It also personalizes training by recalling past interactions, evolving the plan based on user feedback and progress.

The versatility of these custom GPT models lies in their adaptability; criteria can be continuously edited to suit varying use cases and desired outcomes.

Strategic Questions for Implementing AI

It’s important to recognize that the effectiveness of your GPT largely hinges on the industry-specific documentation you provide. Uploading comprehensive and relevant material to GPT Plus significantly enhances the model’s ability to frame responses with accuracy and contextual relevance. Nevertheless, given the online nature of this platform, it is imperative to exercise caution by not including any sensitive company information or material that could contravene corporate policies.

If you are contemplating the development of a bespoke GPT model, the following questions are crucial to consider:

As I write this post, Sam Altman, the CEO and founder of OpenAI, the organization behind ChatGPT, has been ousted from his position by four of the six board members. The reasons behind this decision remain unclear, but The New York Times reported that “The conflict between rapid growth and AI safety came into sharp focus on Friday afternoon, when Mr. Altman was pushed out of his job by four of OpenAI’s six board members, led by Ilya Sutskever. The move has sent shockwaves through OpenAI, the wider tech industry, and its major investors, including Microsoft, which has invested $13 billion in the company.” This event has drawn comparisons from some industry insiders to Steve Jobs’ departure from Apple in 1985, highlighting its significance. As of yesterday, reports emerged that Mr. Altman was in discussions with OpenAI’s board about a possible return.

This story is undoubtedly evolving. The parallel with Steve Jobs and Apple is compelling, especially considering the transformative period Apple underwent during his absence and the groundbreaking innovations, like the iMac, iPod and iPhone, that marked his return. Along a similar vein, I believe Mr. Altman possesses a visionary mindset, crucial in shaping how AI is revolutionizing not only the technological landscape, but also societal norms and the global business economy.

Questions? Please email me here. As always, thank you for reading.
Photo by Yasamine June on Unsplash