Artificial intelligence is all around us. Students are using it. Teachers and staff are using it — and it’s baked in, in some form or fashion, to seemingly every familiar application, from Google Search (Gemini) to Microsoft (Copilot).

A recent Digital Promise survey found that 41 percent of K-12 schools purchased A.I.-enabled education technology in the last year.

But, while A.I. is seemingly everywhere at all once, research suggests schools have been slow to adopt formal usage policies. While most districts (75 percent) offer some form of professional development around artificial intelligence, the same Digital Promise survey found that just 25 percent of schools have specific usage policies.

This is frustrating for a number of reasons, but, given the speed at which the technology is moving, also understandable, says Wall Street Journal Best-Selling Author of Adaptable and K-12 A.I. expert A.J. Juliani. “Any time you develop a policy for something that is continually changing, you’re going to find yourself in a tough spot. Most of the school leaders that I know have waited for state guidance, normally from the school board association, or something along those lines. Some of those have come out. But they are very gray.”

But A.I. is here. And it’s not going anywhere. So, what’s a school leader to do?

While you’re waiting on official guidance from your state, your state board, or wherever, A.J. suggests taking a hard look at three key areas of potential risk to help promote effective and responsible use in your schools.

#1 Safety & privacy – It is critically important that school leaders take necessary precautions, and do due diligence to protect both students and staff. We want to avail of the benefits of artificial intelligence, but not at the expense of our personal security. That means carefully reviewing software privacy policies. Look closely at any data collection, use and sale provisions. It might seem obvious, but more than a few school leaders have landed in hot water for overlooking the details. Make sure personal data is secure and protected. If these stipulations aren’t made immediately obvious, Juliani says it’s best to steer clear of the application.

#2 Ethical considerations – More of a gray area when compared with clearly written safety and privacy policies. You want to ask, how is this A.I. being developed, how is it learning, where is the training data coming from, and what is it doing with the information my staff and students are feeding it? This gets complex. But, “there’s a lot of great resources to help with this,” says Juliani. For instance, Common Sense Media features a host of resources. Spend time evaluating different technologies and applications, have conversations, and don’t be afraid to ask the developers questions about their technology, its use cases and its potential applications for good (and bad) in schools.

#3 Pre-existing usage in schools – When we think about integrating new technology, the question is often do we want to buy this application and implement it in our school or not? But things aren’t so cut-and-dried when it comes to A.I., says Juliani. In many cases, new A.I. is being baked or grandfathered into common applications that students, teachers and staff are already using. “Are you a Google Apps for Education school? A.I. is built into the Google Apps for Education products now. Are you an Apple School? Apple is building A.I into its products now. Microsoft, same thing,” says Juliani. “In many cases, technology that is already approved by your school or district now has A.I. baked into it. It’s important to look at it through that lens first, instead of worrying about, say, kids going to ChatGPT.” Are you looking at the approved applications that you’re currently using for any material changes?