I Have Been "Chatted"
Written by Steve Tadeo, Graphics by Julia Blazynska (LinkedIn)
There, I was blogging about Data Literacy. Getting a few articles posted. I'm feeling pretty good and working on a new article, and suddenly, “Move over, Steve” ChatGPT is here. ChatGPT “Anything you can blog, I can blog better, faster, more thoroughly researched, more viewpoints …………..” My initial reaction was, “I’m done.” My blogging career is over, perhaps not today, but it’s coming.
As a Data Practitioner, I have told people for years, don’t worry; AI will not take your job; it will enhance your job. Make what you do easier and free you up for more complex activities. But then I had this painful realization. Chat is good. Really good.
Steve on Data Literacy:
“Data Literacy is the ability to view data in context, analyze, interpret, decide, and communicate that decision to others.”
ChatGPT on Data Literacy:
Data literacy refers to the skills and knowledge needed to work with data. It involves understanding how to effectively find, interpret, evaluate, and communicate data.
Steve, can you tell me more?
After a few days, "Data Literacy is not binary but rather is on a continuum. Everyone has basic data skills; others are adept at viewing, understanding, and incorporating data into their daily practices. But as the business data world evolves, these basic data literacy skills also continue to grow.”
ChatGPT, can you tell me more?
Seconds later. Data literacy refers to reading, understanding, creating, and communicating data as information. It involves the skills and competencies necessary to effectively collect, manage, analyze, and interpret data. Data literacy is about understanding numbers and statistics and using data to solve problems, make decisions, and communicate insights to others.
Data literacy requires an understanding of data types, data sources, data management, and data analysis techniques. It also involves critical thinking skills to evaluate data's accuracy, relevance, and limitations. In today's data-driven world, data literacy is becoming increasingly important for individuals and organizations to make informed decisions and stay competitive.
Seems clear that Chat will win this challenge. But now what? ChatGPT still needs to be provided a prompt or question to initiate a response, so we are partners for now. Large Language Models will continue to influence our lives, but that level of influence will evolve. I break this evolution into three Phases.
Phase 1 Early Adopters
It took Netflix 5 years to get a million users. Chat did it in 5 days. This is moving so fast that I’m not sure what “Early” means, but the speed at which this is moving creates a succession of 5-day windows. In phase 1, everyone will scramble to find ways to leverage Chat. How do you use Chat to create value and safely, responsively, and accurately? Chat makes mistakes. Businesses will search for competitive advantages; students will leverage Chat to do homework and take tests. The frenzy to use Chat will uncover many good and bad applications, but the foundation for Phase 2 will be set. Chat can help us do what we do faster, with increasing levels of quality and, hopefully, accuracy.
Phase 2 Transformation
Jobs will change, and tasks within jobs will be eliminated. There won’t be an area that will go unaffected, but customer service, legal, and translation work may be hit particularly hard. Everyone essentially will get an assistant or an intern who will become increasingly more competent and require less guidance and support. Lots of efficiencies here, and this surplus will be redirected to new opportunities. Overall productivity will increase dramatically. This transformation phase will lead to the next phase. Where Chat starts operating independently.
Phase 3 Breakthrough
I don’t know what this phase will look like, but it will exponentially differ from Phase 2. We will be faster, smarter, and more productive. Now what? What is on the other side of improvement? Change. What will we do that we haven’t thought of before Chat? The very definition of work will be completely different in Phase 3. What is a workday or workweek? The long-established 8-hour day and 40-hour workweek has been around for about 90 years. Will the workweek be 10 hours? Value creation will accelerate, but the need for “workers” will diminish. Perhaps we will have wealth and time, which are two commodities that are often opposing. What happens when the world when all the data that drive LLM’s are built on data that LLM’s created. The final level will have models’ operation for the most part autonomously. Much of the activity will have minimal or no human intervention required.
Phase 3 is where Chat will replace or, at the very least, radically change my blogging efforts. I won’t be needed for concept generation or execution. So, going forward, I will embrace LLM and Chat. The Genie is not going back into the bottle. Utilize my assistant for collaboration. New ideas, confirm existing and help generate new topics. Continue to explore new areas of application. “Embrace” Phases 1 and 2 and “Brace” for Phase 3
A final note and perhaps a word of caution. Language is the unique characteristic distinguishing humans from every other species on the planet. Other species can communicate, but crafting complex messages is only human. The ability through language to create consensus, group think, coalitions, and shared beliefs was only left to us humans. Politics, Government, Religion, and social media cannot happen without language. Our lock on this capability is slipping away, and at some point, Large Language Models will be developed from “Language” created by Large Language Models. No one knows what this world will look like, but I feel we will find out soon!