Astriata logo mark in orange

A New Frontier in Artificial Intelligence

Illustration of a digital sunrise over a digital landscape with geometric clouds
January 12, 2023

The buzz is infectious—and it’s surrounding something real this time.

It’s the recent developments in artificial intelligence (AI), making the front pages of major newspapers and filling up the airways of tech podcasts everywhere.

Pundits compare the latest AI developments to everything from the advent of electricity to the beginning stages of the internet, promising a game changer that is now in its infancy.

The big question among our team at Astriata is: What do the AI developments mean for those of us who work in the association, nonprofit, healthcare, government, education, and B2B spaces? How can we use them to work more efficiently and effectively, without sacrificing our ethics and integrity as we labor to achieve humanitarian-minded goals? After all, our use of AI often looks different than it does for mega-corporations.

This month, we’re diving in to explore and bring you helpful information on what the new developments in AI potentially mean for your day-to-day and longer-term business operations. We’ll also explore some of the ethical issues, risks, and controversies surrounding the new AI, while sharing tips on what you can do to start using the AI tools at your disposal now. To kick off our series, we’ll explain the recent developments in the context of the larger AI world, while pointing out some of the ways you can use them to your benefit.

Understanding generative AI

For starters, it’s important to understand the term generative artificial intelligence, which you’ve likely encountered lately in your news feeds. Generative AI utilizes algorithms to sort through and pull from large data sets to generate (or produce) new data that aligns with what is already out there. It’s a form of machine learning that involves identifying patterns in data and then using those patterns to create something “original.” (Note that we put “original” in quotation marks because what is generated is actually a synthesis of what already exists in the data set. But it is, in fact, based on what the computer learns and presented in a novel way. In this sense, it is original.)

How does this differ from the AI we’ve seen and used for the last five to ten years? In large part, existing AI has been task-specific, while the new AI utilizes “models that are trained on a broad set of unlabeled data that can be used for different tasks, with minimal fine-tuning,” explains Mike Murphy, the editor in chief of IBM’s research blog.

In other words, the new AI can use what it learns to perform a broad range of new tasks (much like a human). It’s based on what the industry refers to as a foundation model, a term coined by Stanford University’s Institute for Human-Centered Artificial Intelligence. “Train one model on a huge amount of data and adapt it to many applications,” the Institute explains on its homepage. “We call such a model a foundation model.”

That first model, then, serves as the foundation on which the model and generative AI stand. “As the name suggests, [a foundation model] can be the foundation for many applications of the AI model,” Murphy of IBM says. “Using self-supervised learning and transfer learning, the model can apply information it’s learnt about one situation to another.”

Murphy compares foundation models to the experience of learning to drive a car. “You learn to drive on one car, for example, and without too much effort, you can drive most other cars—or even a truck or a bus,” he says. This is an enormous development in AI—one that experts say has the potential to help us immensely and even some of the world’s most complex problems, from climate change to cancer.

What, though, does it mean for your organization?

Enter: ChatGPT

Let’s start with the tool everyone’s talking about: ChatGPT, the chatbot released at the end of 2022 and now available for free to the general public. (If you’re not familiar with the term chatbot, know that chatbots are computer programs designed to simulate conversations with human users, typically over the internet.) Built by OpenAI, ChatGPT allows users to enter a question or command, and then the chatbot pulls from its data (and what it’s learned from the data) to generate an original answer.

Say, for instance, you’re a healthcare organization, writing a letter to potential donors about the importance of funding diabetes research. You can enter your command in ChatGPT:

Write a letter to donors about the importance of supporting diabetes research.

And ChatGPT will draft an actual letter in a matter of seconds.

We tried it out, and here’s what the chatbot generated:

Dear donors,

We are writing to thank you for your ongoing support of our efforts to find a cure for diabetes. As you may know, diabetes is a serious and chronic condition that affects millions of people around the world, and we believe that with your help, we can make a real difference in the lives of those living with this disease.

Your support is more important now than ever, as diabetes rates continue to rise around the globe. By funding research into new treatments and potential cures, you are helping to ensure that people with diabetes have access to the best possible care and can live healthy, productive lives.

We are grateful for your generosity and commitment to this cause, and we hope that you will continue to support our efforts to find a cure for diabetes. Thank you for all that you do.

Sincerely,

[Your name]

It’s not a perfect letter, but it’s also not bad. At the very least, it gives the letter writer some ideas to work with as they go about creating a draft.

But ChatGPT can do far more than write letters. The chatbot can dig up statistics for your proposal or report or draft part of your grant or essay. It can write computer code, make a guess at a medical diagnosis, explain complex concepts, generate personalized responses to customer inquiries—the list goes on.

ChatGPT can even tell jokes and be creative. For example, when we asked it to tell a joke about the marketing industry, here’s what the chatbot generated:

Why was the marketing company’s office always cold?

Because they left their windows open for new ideas.

Kinda funny, eh? We think it’s a good one.

To get a sense of ChatGPT’s potential, we encourage you to create a free account and play around. The plus side is that it’s easy and straightforward to use, with no long-winded tutorials required.

We’d love to hear how it goes—let us know on socials. And stay tuned for more to come on ChatGPT and other recent developments in AI, including the groundbreaking image generator DALL-E 2.

If you missed it before the holidays, read our article about another important development in technology, Google Analytics 4 (GA4).


We're here for you.