III. The Current State of Artificial Intelligence in Education

Artificial intelligence is a rapidly growing technology that is actively changing how we teach, learn, work, and live. AI can be found nearly everywhere. Many people knowingly and unknowingly interact with AI daily, from mobile phones and computer applications to surveillance systems and media streaming services. In its most basic definition, artificial intelligence is any machine-based system designed around human-defined objectives to perform tasks that would otherwise require human or animal intelligence. Varying in purpose and programming, AI systems can do one or more of the following: identify patterns, understand natural language, produce content, make predictions and decisions, give recommendations, solve problems, or adapt to and learn from new information and circumstances.

In general, current artificial intelligence tools can be broken down into three categories:

Reactive AI tools respond to specific inputs or situations without learning from past experiences—for example, AI assistants, such as Alexa and Siri, or household tools, such as Roomba vacuums.
Predictive AI tools analyze historical data and experiences to predict future events or behaviors, such as when Amazon or Netflix shows you suggested items.
Generative AI tools, such as ChatGPT and Gemini, generate novel text, images, videos, or other content based on existing data patterns and structures.

While generative AI has received the greatest attention in recent months, it is important to realize that humans have been using reactive and predictive AI tools for quite some time. In addition, AI technology is developing very quickly, with new models and tools introduced frequently.

Artificial intelligence employed in pre-K–12 and higher education contexts can take on a variety of forms. A report by Education International provides a helpful construct for these uses:

Student-Focused AI includes adaptive tutoring systems, automatic writing evaluation systems, and chatbots, among other tools aimed at supporting students.
Teacher-Focused AI tools are aimed at teachers and include assessment supports, lesson planning tools, and resource curation systems.
Institution-Focused AI helps with school and campus administration and operations, such as handling scheduling, scanning for safety concerns, and identifying students at risk.

A fourth type, system-focused AI, has also begun to emerge, with some states using AI to determine school funding or score state assessments.

Recognizing this power, capability, and financial opportunity, technology companies and developers are actively finding ways to integrate AI into education systems worldwide. Yet, at this point, many uses of AI in education are largely speculative, without a strong, independent research base showing that these tools are more effective than existing practices or technologies.

Nonetheless, students and educators have started to embrace artificial intelligence, particularly generative AI. A 2024 report by the Center for Democracy & Technology found that the percentage of K–12 teachers who reported using a generative AI tool for personal or school use jumped 32 percentage points, to 83 percent, between the 2022–2023 school year and 2023–2024. In the same study, 59 percent of teachers reported that they are certain at least one of their students has used generative AI for school purposes. In higher education, 49 percent of students reported using generative AI regularly as of September 2023, although only 22 percent of faculty reported this level of usage.

This surge in artificial intelligence, and particularly in generative AI, requires that educators become prepared to assess when it is appropriate to use AI, help their students become AI literate, and advocate for the development of policies about this technology. In other words, educators must be able to not only teach with AI but also teach about AI. Yet, opportunities for educators to get up to speed are still lacking. In a survey taken earlier this year, Education Week found that 71 percent of K–12 teachers had received no professional learning about using artificial intelligence in the classroom.

As of this writing, only 16 states—Arizona, California, Connecticut, Hawaii, Indiana, Kentucky, Michigan, Mississippi, North Carolina, Ohio, Oklahoma, Oregon, Utah, Virginia, Washington, and West Virginia—have issued guidance from the state department of education or another organization. Meanwhile, Tennessee has mandated that districts develop their own policies. New York has issued a statewide ban on the use of facial recognition in education settings. Districts and higher educational institutions have also varied greatly in their approaches, with some banning AI outright (although, some of those bans have since been lifted), some putting policies in place about appropriate use, and some, like the Los Angeles Unified School District, building AI-powered platforms. Education systems are clearly in a transitional phase in terms of determining when and how to harness AI.

Navigating this significant technological shift will require intense attention and involvement by the NEA, its state and local affiliates, and its members. Members and affiliates need to be prepared to be leaders at their schools and campuses and in policy discussions at all levels of the education system. The needs of students and educators must be at the forefront during the development, selection, implementation, and evaluation of AI technologies to ensure that these tools support effective teaching and learning, not the agendas of for-profit entities or those who would like to undermine public education by replacing school staff with computers. We must also be ready to hold AI developers accountable to protect data privacy and intellectual property rights, mitigate algorithmic bias and inaccurate or nonsensical outputs, and diminish environmental hazards.

This report provides background to the proposed Policy Statement on the Use of Artificial Intelligence in Education by providing an overview of the promise of AI in education, reviewing existing NEA policies, and providing background research and information on each of the five principles to support the safe, effective, and equitable use of AI technologies in schools and on campuses. The Task Force acknowledges that AI is developing and changing at a rapid pace, and thus, policies must be adaptable and reviewed regularly. Our goal for the Policy Statement and this accompanying report is to provide a starting point for an ongoing conversation about how artificial intelligence should and will become a part of education and society.

3 The Glossary in Appendix B provides a definition of this and other technical terms.
4 Adapted from ASCD et al., Bringing AI to School: Tips for School Leaders (2023), https://cms-live-media.iste.org/Bringing_AI_to_School-2023_07.pdf.
5 Wayne Holmes, The Unintended Consequences of Artificial Intelligence and Education (Education International, 2023), https://www.ei-ie.org/en/item/28115:the-unintended-consequences-of-artificial-intelligence-and-education.
6 Mark Lieberman, "A State Uses AI to Determine School Funding. Is This the Future or a Cautionary Tale?," Education Week (February 28, 2024). https://www.edweek.org/policy-politics/a-state-uses-ai-to-determine-school-funding-is-this-the-future-or-a-cautionary-tale/2024/02.
7 Keaton Peters, "Texas Will Use Computers to Grade Written Answers on this Year’s STAAR Tests," Texas Tribune (April 9, 2024). https://abc7amarillo.com/news/local/texas-will-use-computers-to-grade-written-answers-on-this-years-staar-tests-texas-education-agency-assessment-of-academic-readiness-jose-rios-chris-rozunick-gpt-4-googles-gemini-ai-carrie-griffith.
8 Holmes, The Unintended Consequences of Artificial Intelligence and Education.
9 Maddy Dwyer and Elizabeth Laird, Up in the Air: Educators Juggling the Potential of Generative AI with Detection, Discipline, and Distrust (Center for Democracy & Technology, 2024), https://cdt.org/wp-content/uploads/2024/03/2024-03-21-CDT-Civic-Tech-Generative-AI-Survey-Research-final.pdf.
10 Tyton Partners, GenAI in Higher Education: Fall 2023 Update Time for Class Study (2023), https://tytonpartners.com/app/uploads/2023/10/GenAI-IN-HIGHER-EDUCATION-FALL-2023-UPDATE-TIME-FOR-CLASS-STUDY.pdf.
11 Lauraine Langreo, "Teachers Desperately Need AI Training. How Many Are Getting It?," Education Week, March 25, 2024, https://www.edweek.org/leadership/teachers-desperately-need-ai-training-how-many-are-getting-it/2024/03.
12 For an overview of current AI policymaking in higher education, see Tyton Partners, GenAI in Higher Education: Fall 2023 Update Time for Class Study.
13 "Ed Powered by Individual Acceleration Plan," Los Angeles Unified School District, 2024, http://www.lausd.org/site/default.aspx?PageID=19406.
14 Some use the term ‘hallucinations’ to refer to when AI tools provide nonsensical or inaccurate outputs. We do not use this term because it anthropomorphizes artificial intelligence and is insensitive to people with mental illness.