For more than a decade, we’ve barreled through educational curriculum development across the country (and in many parts of the world) seeking “more tech, more tech, more tech in school.” Technology was supposed to democratize education, make it available to the masses, enhance engagement, and make the whole educational process more effective.

students in classroom at desks using tech in school — Image via Canva

As studies with ten-plus years of data showing declining academic performance (even before pandemic declines) and questionable effectiveness of technology use in classrooms begin to pile up, the promise of a tech-first or tech-forward education is coming under serious scrutiny.

For years, many people (especially parents) have intuitively known that something about the tech doesn’t feel quite right. Interestingly enough, the very people who lead the companies that make all this “great educational technology” often send their children to schools that don’t use it. Yet they tout its benefits to anyone with influence over school tax dollars who can put the technology in front of children other than their own. 😏 I’m not impressed…

Dissent against tech-first classrooms is gaining steam around the country. Locally, a growing group of UCF parents, led by the grassroots group SlowTech UCF (full disclosure: I’m part of this group), presented a multi-pronged request to the district at the May 18, 2026, school board meeting, requesting that the administration revisit various elements of the tech-first curriculum design in our school district.

Twelve parents, including me, spoke about specific concerns. Ten of us specifically worked together as part of Slowtech UCF and presented a coordinated argument to the school board, asking for changes to school policies and practices, including but not limited to:

Ensuring the default mode for teaching is based on human instruction and not rooted in technology
Limiting screen time (amounts depending on grade levels)
Prohibition on screen time as a reward for good behavior or as an activity available during indoor recess, lunch, and other periods of downtime
No open internet access on school devices (eg, updated policy would allow students to access a particular YouTube video as needed, but would not allow unlimited access to YouTube generally)
Moratorium on gameified tools (like Prodigy, Blooket, GimKit, DuoLingo, and others) as these reinforce tech addiction habits, and have unclear educational value relative to non-tech alternatives
Public disclosure of evidence standards, conflicts of interest, and performance claims related to district purchases of EdTech tools.
Assurance that student data isn’t being collected or sold by third-party vendors
Enhancements to parental controls to allow parents more influence over how school-issued devices are used, especially at home
Additional modifications to the current policy around cell phones in schools
Moratorium on generative AI and use of large language models on school-issued devices, at a minimum for grades K-8, given that the terms of use for AI platforms prohibit consent by users under the age of 13. Claude and ChatGPT require parental consent for users under age 18, and parents should have an avenue to withhold consent for AI use.

To support these requests, the group provided a variety of handouts to the board, including:

scientific studies and data
testimony provided to the United States Senate by Dr. Jared Cooney Horvath (written testimony here)
summary of public data on the declines in UCF student performance over the last ten years (friends, it’s not great… 👀)

Looking back at UCF standardized test scores from 10 years ago (before 1:1 Chromebooks) to this past year, at students who scored proficient or advanced:

PSSA ELA scores decreased by 12% on average.
PSSA MATH scores decreased by 5% on average (17% in 5th grade).
PSSA SCIENCE scores are down 3% in 4^th grade and 10% in 8^th grade.
On the Keystones tests, testing scores declined by 21% in Algebra 1, 17% in Biology, and 9% in English.

The pandemic played a role in some of this, and correlation doesn’t prove causation. But this isn’t a good look for the salespeople encouraging the district to spend more of our tax dollars on their software and devices (especially when their kids attend tech-free, phone-free schools).

The changes SlowTech UCF requested range from fairly easy to significant overhauls. Hopefully, the district can start implementing low-hanging fruit soon while dedicating meaningful resources to investigating more robust reductions in tech-dependent curriculum, based on growing research and data.

I encourage you to watch the May 18, 2026, school board meeting to see the comments and concerns in context. Public comments from Slowtech UCF members and others speaking on tech use in school start just after minute 42:09.

What does AI think about tech in school?

So let’s play a fun game and ask the techiest of tech if tech in school is good for students and academic outcomes.

I posed this question to Claude, Anthropic’s LLM chatbot:

Based on the last 10-20 years of educational data in the United States, do you think that technology in the classroom is helping or hurting how students learn and how effective school curriculums are?

I think you’re gonna love Claude’s response:

—- Quoted from Claude —

“This is one of education’s most genuinely contested empirical questions, and the honest answer is: the evidence is mixed, and it depends heavily on how technology is used.

Here’s a fair breakdown of what the research actually shows:

Where technology appears to help

Personalized pacing — adaptive learning software (like Khan Academy or IXL) lets students work at their own level, which benefits both struggling and advanced learners more than one-size-fits-all instruction.
Access and equity — for rural or under-resourced schools, technology can expand access to AP courses, specialist instruction, and enrichment that simply wouldn’t exist otherwise.
Engagement with specific populations — students with learning disabilities often benefit from assistive tech (text-to-speech, speech-to-text, organizational tools).
Immediate feedback — digital tools can give students real-time corrections in ways a teacher managing 30 kids can’t.

Where technology appears to hurt

Reading and comprehension — studies consistently show that students retain less when reading on screens vs. paper, particularly for complex texts requiring deep analysis. Norwegian and French national studies found measurable drops in reading comprehension tied to device adoption.
Math fundamentals — over-reliance on calculators and apps before procedural fluency is built tends to produce fragile understanding.
Attention and working memory — device multitasking is well-documented as harmful to learning. Even the presence of a phone (not in use) measurably degrades cognitive performance in classroom studies.
NAEP score trends — U.S. National Assessment of Educational Progress (the “Nation’s Report Card”) scores in reading and math have largely stagnated or declined since 2012, the same period during which 1:1 device programs became widespread. Causation isn’t proven, but the correlation is uncomfortable.
The pandemic remote-learning collapse — COVID made clear that screen-dependent instruction is a poor substitute for in-person learning, with the most vulnerable students suffering most.

The nuance that matters most

The research strongly suggests technology is a tool multiplier — it amplifies whatever teaching practice surrounds it. Good teachers using tech thoughtfully see benefits. Weak pedagogy + devices = worse outcomes than weak pedagogy alone, because devices add distraction without adding structure.

Countries that perform best on international assessments (PISA) — like Singapore, Estonia, and Japan — use technology selectively and deliberately, not as a default. The U.S. trend of “more devices = better education” has not been validated by outcomes data.

Bottom line

The weight of the last decade’s evidence leans toward technology having been oversold and undermanaged in American schools — deployed faster than teachers were trained to use it, and often substituting for rather than enhancing instruction. That doesn’t mean technology is bad in classrooms; it means implementation quality matters enormously, and the U.S. hasn’t generally gotten that right at scale.

The most defensible position the data supports: targeted, structured, teacher-directed use of specific tools (not open device access) improves outcomes; unrestricted or passive device use generally doesn’t.”

— End Claude conversation —

All of the information above needs validation before it is used to inform broad curriculum modifications. I did not fact-check this information; that wasn’t the point. But I think our intuition reaffirms what Claude suggested.

Even Big Techie Tech thinks tech in school is mostly unhelpful in the United States, and certainly unhelpful when applied without highly targeted intent. At UCFSD, the entire curriculum is on Chromebooks (at least once you get to middle school). Students don’t have textbooks and do much of their work directly on the computers. Tech in school at UCFSD is applied with a broad brush rather than through highly targeted touchpoints.

How is UCFSD using AI in the curriculum?

As it relates to tech in school, the UCFSD AI train really hasn’t left the station yet. Kids and teachers use AI without much guidance (at least so far), but it’s not widely integrated into the formal curriculum. Some AI functions on the school-issued devices are blocked, while others remain available to students, with limited guidance so far on how to use them ethically or effectively.

The district approved its first AI policy in March 2026. It’s high-level guidance, with a red/yellow/green framework, to support deeper implementation by the administration as the technology crops up in various ways throughout the school day.

However, I’m concerned about how it will be incorporated into the curriculum going forward. We don’t know which uses of AI in school will prove fruitful and which will be harmful. Jonathan Haidt, a longtime advocate for unburdening childhoods from excessive technology, recently introduced a three-prong approach to techno-skepticism for children. I encourage you to watch his TED Talk on the matter.

If reading is more your jam than watching videos, he wrote this piece that corresponds with the video. Take a look to learn more about his “three general principles that would reduce the harm to children from current and future technologies:

Beware of artificial relationships for minors.
Protect brain development through puberty.
Prioritize people and books in education, not screens.”

Will AI make us dumber?

Each of the SlowTech representatives covered different matters related to technology in the classrooms and schools’ curricula. I addressed AI broadly in the context of the philosophical purpose of education and AI’s potential impacts on cognitive impairment.

If you’re curious, here’s what I shared at the school board meeting:

I’m Jen Panaro, parent of two Patton MS students in 6th and 8th grades. I live in Chadds Ford Township. As someone who uses AI almost daily, I’m concerned about how it might be implemented in school and compromise the goals of education and development.

There’s endless discussion swirling around AI and how it’s going to change everything. We must learn how to use AI to stay relevant and succeed. Most of these conversations revolve around how AI is used in our production markets: businesses, non-profits, and other organizations whose mission is to provide products or services.

As an educational institution, the mission is wholly different. For a business, using AI to expedite the delivery of its deliverables can help achieve its mission more efficiently.

In school, however, the deliverable is not an end product or service. The deliverable is the process of learning itself. AI and many tech applications and devices tout how they reduce friction and enhance productivity. But in school, the friction is the learning. The friction builds endurance to tolerate uncertainty, new ideas, different opinions, and challenging circumstances. What is creative or critical thinking if not sitting in the friction and discomfort of the unknown to explore what’s possible and create something new?

We’ve succeeded in education when we have taught our students how to muddle through the messy middle. Racing to the end to write a paper more efficiently by letting AI handle outlines, idea summarization, or analysis doesn’t achieve the mission more efficiently. Instead, it skips over the mission entirely.

LLMs and generative AI can be incredibly helpful when used effectively. They also provide a lot of shortcuts that, frankly, might make us dumber. When we let AI do the cognitive work of learning, those cognitive muscles atrophy instead of develop. In the business world, that might expedite the mission. At school, it’s counter to the mission.

I hear that kids “need to learn how to use AI.” These tools are not hard to use; they’re designed to be incredibly intuitive. Let’s be honest: kids are going to learn how to use AI outside of school. These tools will be so pervasive that the most successful schools might be the ones that teach students how to do what the AI-obsessed world overlooks.

How will students learn how to craft new ideas if they use LLMs to help them draft an outline first? How do students learn to deeply engage with texts in meaningful ways if they rely on AI summaries and assume that LLMs have distilled all of the relevant ideas properly? Won’t AI, by definition a regression to the mean, generally suggest the same final analysis to all of us, leaving little space for multiple interpretations?

I know the school has a red/yellow/green framework for using AI. I’m not that concerned about the red or green categories. But I’m very concerned that too many tools will wind up in the yellow category, ultimately diminishing students’ cognitive muscle-building.

The collective seems to be concerned that we will all become obsolete if we don’t know how to use AI. I’m concerned we will each become obsolete if we don’t know how to be human without AI.

I’m not suggesting there’s no place for LLMs or generative AI in a well-designed curriculum. But I do think we need to take a very thoughtful approach, where slow and steady wins the race.

What do you think about tech in school?

There’s no definitive answer on how AI should be incorporated into the educational curriculum because it’s too early to know. But I’d love to know what you think? Our school board and district administration have some tough choices to make about how to incorporate technology into the school day or untangle it from it. How do you hope they move forward with technology integration?

If you’re interested in joining the SlowTech movement, head to the website for more information. Signing up for the newsletter is the best way to stay up to date on what the group is doing and to jump in when you find something that resonates with you!