Empowering K-12 Districts: Navigating AI Adoption for Enhanced Educational Efficiency and Effectiveness

Introduction

What is Artificial Intelligence (AI)?

Drawing from disciplines including computer science, math, philosophy, and cognitive science, AI refers to the field of research and practice that seeks to produce technological innovations that replace manual work. Its aim is to invent machines and computer programs that can mimic human functioning, or complete tasks which typically require human intelligence (Anjila, 2021).

A Brief History

Although the present AI mania makes it seem as though new technologies, like OpenAI’s ChatGPT, hatched without warning, researchers began building their foundations nearly two centuries ago. Twentieth-century mathematician and computer scientist, Alan Turing, is often credited as the father of AI, however its seeds were planted in the 1830’s when Charles Babbage began developing blueprints to create The Analytical Engine. This steam-powered Industrial Revolution invention would become the first programmable computer.

If built according to plan, the machine would be able to perform calculations, alter them while in progress, and store them in its limited memory. The thought of a machine capable of performing processes previously reserved for human cognition set off alarm bells for many. In 1843, Ada Lovelace, a mathematician who collaborated with Babbage on the project attempted to quell public fear, writing: “The Analytical Engine has no pretensions to originate anything. It can do whatever we know how to order it to perform.”

Fast forwarding nearly two centuries shows that while technology has evolved beyond recognition, the AI agenda and conversation have hardly changed. Echoing Lovelace, contributing writer to The Atlantic, Ian Bogost (2022) reasoned on generative AI:

ChatGPT lacks the ability to truly understand the complexity of human language and conversation. It is simply trained to generate words based on a given input, but it does not have the ability to truly comprehend the meaning behind those words. This means that any responses it generates are likely to be shallow and lacking in depth and insight.

In between Babbage and ChatGPT, well before generative AI became our reality, it became a thread in our cultural tapestry. Literature, television, and movies explored the philosophical and ethical implications of AI. In his 1899 publication, The Wonderful Wizard of Oz, L. Frank Baum tinkered with the idea of implanting human cognition into men made of tin and straw. His scarecrow considers:

‘It must be inconvenient to be made out of flesh…for you must sleep, and eat and drink. However, you have brains and it is worth a lot of bother to be able to think properly.’ (Baum, 1899, p. 56).

In the second half of the twentieth century, advances in research, funding, and technology actualized these science fiction fantasies. These advances likely piqued AI public discourse and interest which is illustrated by the data displayed in Figure 1. It visualizes the frequencies with which the term “artificial intelligence” appears in books published between 1950 and 2019. Notice that there are two peaks: one in the 1980’s and another in the late 2010’s.   

Figure 1 – Google Ngram Viewer for Keywords “Artificial Intelligence”

The AI Timeline 

See the timeline to trace the AI conversation through the years.

AI Today

Today, AI machines and programs have infiltrated our lives and world. Rule-based AI, or systems that operate on a set of pre-defined rules created by human experts, assist with a variety of tasks. Robot vacuum cleaners autonomously clean our floors; back-up cameras ensure that nothing is in the way when we shift our vehicles into reverse; and facial recognition software securely locks our phones until they scan our facial features.

Predictive AI includes systems that can sense, comprehend, and respond based on historical data. Auto-complete predicts the next word or words in a text message, email, or search query while auto-correct identifies misspelled words and replaces them with correct spellings. Online shops recommend products based on browsing or purchasing history.

Consumers have largely welcomed and come to depend upon these advances, especially those that have assumed repetitive and tedious tasks that many were eager to delegate. For many users, AI functions and effects are now so ingrained that they have become invisible. For instance, an analysis of case studies of individuals’ awareness of AutoCorrect’s changes on their writing found that some users vastly underestimate the number of times the AI intervened (Wood, 2014).

However, attitudes towards AI have shifted more recently as rapid technological advances paved the way for Generative AI, or systems that simulate human-like cognitive processes to create new content. Large language models, like OpenAI’s ChatGPT, which are computer programs that can understand and generate content including language, have become the focus of the AI conversation. To learn common human language patterns, large language models were trained on a diet of internet texts including Wikipedia pages, Reddit conversations, and published books, though the exact texts have not been disclosed (Brown et al., 2020). Based on their training, these models calculated the probabilities of words appearing next based on a given context. To generate output, they predict the next word based on prompt keywords and previously generated words. These models have brought Turing’s (1950) AI aspirations to life, performing as well as and better than humans on a range of standardized assessments (see Figure 2).

Figure 2 – GPT Performance on Common Standardized Assessments

Their demonstrated utility and accessibility has made large language models, like ChatGPT, highly popular. It drastically outperformed (2 months) other applications like, Instagram (2.5 years), Facebook (4.5 years), and Twitter (5 years), in the race to gain 100 million users. See Figure 3 for a comparison of ChatGPT’s adoption rate to that of other popular applications.

Figure 3 – The Time it Took for Popular Online Services to Reach 100 Million Users

Despite the remarkable capabilities of these emerging technologies, Americans remain ambivalent toward them. Surveys by the Pew Research Center (see Figure 4) reveal a shifting sentiment: in 2021 and 2022 most US adults felt a balance of excitement and concern regarding the growing influence of AI in their daily lives. However, by 2023, the scales had tipped, with a majority expressing more apprehension than enthusiasm. Among reasons for concern include job displacement, heightened surveillance, lack of human connection, and the increasing power of AI.

Figure 4 – Pew Survey Results: What the Data Says about Americans’ Views of AI