How to Get a Quick Win with GPT-4

The most common question Prolego receives from analytics and data leaders in large enterprises is, "What GPT use cases should we consider?"

I'm assuming you're seeking a "quick win" like everyone else. You need a quick win to demonstrate feasibility, educate your business partners about the technology, and build organizational momentum. Here are some suggestions for identifying use cases and selecting the best candidates for a quick win opportunity:

  1. Start by asking ChatGPT. Although it may not seem like an inspiring answer, many have found it a great starting point. Your business has common processes with other companies in your sector, so use cases tend to be similar. For instance, if you work in insurance, your key operations likely involve underwriting, claims, support, and distribution. Spending 15 minutes with ChatGPT can generate five to ten possibilities.
  1. Consider data constraints. Many projects struggle due to data issues, so identify projects with accessible data, minimal security or privacy concerns, and cooperative partners. Data constraints are often organizational constraints, with uncooperative data "owners."
  1. Find a motivated business partner. No matter how well-executed, projects will fail if your core business constituent lacks motivation. Disruptive technologies like GPT-4 require business customers to change how they work, so it's crucial to have a partner who shares your vision and goals.
  1. Programmatically-generated training data. Choose a use case that doesn't require significant input from end users for generating training data. Opt for situations where labeled training data can be generated using techniques like weak supervision or embeddings. Aim for use cases where your data science team can develop a proof of concept without substantial investment in interaction between business customers. Avoid use cases where significant user-generated prompt engineering is required. Document Q&A and semantic search are great starting points.

I hope these guidelines prove helpful. Successful analytics leaders invest time upfront to identify the right opportunities. There's no shortcut when dealing with complex new technology and significant changes to organizational processes.

Subscribe to our YouTube channel where we post daily videos on the ever-evolving world of AI and large language models.

PROLEGO GREATEST HITS:

ABOUT PROLEGO
Prolego is an elite consulting team of AI engineers, strategists, and creative professionals guiding the world’s largest companies through the AI transformation. Founded in 2017 by technology veterans Kevin Dewalt and Russ Rands, Prolego has helped dozens of Fortune 1000 companies develop AI strategies, transform their workforce, and build state-of-the-art AI solutions..

Let’s Future Proof Your Business.