Artificial intelligence experts warning the technology may wipe out humanity in the next decade.
A swarm of AI-powered agents pulling off a rogue hack.
CEOs of big AI corporations publicly pleading to put the brakes on technological development.
The headlines about the technology’s potential perils are enough to make an AI-focused teacher professional development session feel like the first scene in a dystopian science fiction film.
AI can already be used to impersonate a principal’s voice or help a hacker breach schools’ cybersecurity safeguards. If AI bots could indeed unwittingly, or upon a human order, release a deadly virus, launch a nuclear weapon, or wreak havoc on vital infrastructure, as some experts warn, should a teacher think twice before using a chatbot to craft parent emails?
Will heightened fear of AI fuel the growing parent pushback to the overuse of technology in schools?
There’s already a lot to worry about when it comes to AI in K-12 education—student data privacy, cognitive offloading, kids creating deepfakes—for educators to waste mental energy on some theoretical, apocalyptic future, argued Justin Reich, the executive director of MIT’s Teaching Systems Lab, which researches the future of teacher learning.
“Educators and families have plenty to talk about, plenty to be concerned about, just addressing the harms that are caused today by these systems,” Reich said. “You don’t have to speculate about future harms.”
Reich, and some others, are skeptical of the companies’ motivations in detailing AI’s potential downsides. (Some critics have suggested that the CEOs of big AI companies are calling for regulation of the technologies they’ve created to ward off potential competitors, or showcase the technology’s power.)
“These people are not honest,” Reich said of the tech CEOs. “They’re self-motivated, and they’re really bad at predicting the future.”
In calling for a slowdown in AI development, Dario Amodei, Anthropic’s CEO, said the technology is developing too quickly for engineers to ensure that the requisite safety measures are in place.
The company has “tried to prioritize caution over speed and prudence over profit,” he wrote. But “over the last few months, I have become convinced that fully addressing the risks requires even more prudence—not just investing in risk prevention, but pacing the rate of capabilities advancement so that risk prevention has time to keep up.”
Is slowing down AI adoption in education a bad thing?
Even so, the tidal wave of scary news stories may spur more school districts to consider banning or pausing adoption of AI tools for teaching and learning and the management of schools, experts predicted.
“These stories are truly scary, and when we’re scared, we want to push pause, and we want to protect our children,” said Jeremy Roschelle, the co-executive director of learning science research for Digital Promise, a nonprofit organization that works on equity and technology issues in schools. “I think it will increase the momentum to ban or pause” the use of student-facing generative AI tools in schools.
Florida’s state board of education this month called for districts to allow families to decide whether to “opt-in” to letting their children use generative AI learning tools.
The nation’s two largest districts—New York City Public Schools and Los Angeles Unified—recently hit the brakes on student use of AI tools. The Los Angeles policy restricts use of the technology for all students. New York prohibits it in elementary and middle school but allows limited use in high schools.
This new pressure to move more carefully and deliberately in AI adoption might give districts a chance to catch their breath and decide when and how the tech can be most helpful, said Rebecca Winthrop, a senior fellow and director at the Center for Universal Education at the Brookings Institution, where she led an examination into the potential problems and benefits of generative AI in K-12 schools and concluded the risks outweighed the upsides.
A pause in the AI hype “will make it easier to actually slow down and try to do what I would call very good, useful, narrow AI use,” Winthrop said.
Such “narrow"—but potentially effective use of AI—might include tools that help teachers analyze student data in real time to figure out where students are struggling. Or it could feature use of an older form of the technology—predictive AI, which can identify patterns and make suggestions based on them—to give students personalized reading and math practice problems, she said.
Winthrop also recommended school districts steer clear of agentic AI, a more advanced version of the technology than generative AI. Agentic AI is designed to be proactive and perform multi-step tasks, such as booking travel or coding and fixing software programs.
“They should be very careful of having any agentic AI in their schools because it is not tested,” Winthrop said.
Rising AI fears could prompt district leaders to take a harder look at tech tools before buying them
The panic-inducing AI news may prompt district leaders to take a harder look at ed-tech tools that incorporate the technology, instead of buying hundreds of tech tools with little scrutiny, as some school systems have done in the past, Roschelle said.
“You don’t want a thousand new products entering schools that have been vibe coded and you actually don’t know what’s in them and what they do because they’ve been built very fast with not enough human supervision,” Roschelle said.
That’s what Matthew Callison, the director of innovation and strategic partnerships for the South Fayette Township public schools near Pittsburgh took away from the spate of recent news stories about the destructive powers of AI.
“As educators, we need to be careful that we’re vetting the tools that we’re bringing in,” Callison said. “There are concerns that we need to take into account. There are guardrails we need to have in place.”
At the same time, he added, “we need to be preparing students to understand what the technology is, how to use it to empower themselves, and how to use it responsibly, and have these constructive conversations, helping students think through all the implications and pros and cons,” he added.
For some students—not the youngest—that could involve using generative AI tools, with teacher guidance and support, Callison said.
“Our job is to help prepare students to be successful beyond our four walls, and that involves having conversations with caring adults and helping [students] understand and navigate these tools and these issues,” he said.
Students encouraged to provide input about how to think about the role of AI in schools
That’s the approach Heidi Sipe, the superintendent of the Umatilla School District in Oregon, has taken.
Sipe has asked her student advisory board to help her think through the district’s approach to the technology. Students advocated for keeping AI out of writing classes, other than allowing it for initial brainstorming.
But students also see the technology’s value. For instance, some students shared positive experiences with virtual chatbot tutors, Sipe said. And they’ve suggested steps, such as training parents how to use AI, so they can help safeguard their children’s use of the technology in more meaningful and sophisticated ways.
Last summer, Umatilla students joined a group of 98 high schoolers from around the country to craft a national AI framework. Students from the district have also spoken about the technology’s implications for K-12 with state lawmakers.
The potential dangers of AI will continue to be part of developmentally appropriate discussions of the technology, Sipe said.
“We have to be very cautious and think through what our core work [as educators] should be, and that is to develop strong learners who are skilled in critical thinking,” Sipe said.
“And so, as we’re talking with students about all [this information], they should be reading, and they should be learning and they should be questioning. They should be asking: what role should AI play in my life, knowing that there are things beyond all of our control right now and that’s concerning.”
