Dissident Blog

When Watching Students Becomes the Lesson

By Hoto · October 24, 2026 · Ideas for change

Schools are collecting more data on students than ever. I think they should have to show it helps, explain it, and let people push back when it is wrong.

The everyday practice

A ninth grader types a dark joke into a school-issued laptop at 11 p.m. By morning, monitoring software has flagged it and an administrator has a report. Down the hall, a student's attendance app logs three late arrivals, which triggers an automatic penalty. In some districts, cameras at the entrance have been set up to match faces against a watch list.

None of this looks dramatic. It arrives as software updates, safety grants and dashboard features. That is exactly why it deserves a closer look. The philosopher Michel Foucault argued that modern institutions shape people less through force than through constant observation, records and ranking. People who never know when they are being watched start to police themselves as if they always are. Students learn not only what is taught, but also that someone is always watching and scoring them.

This is not a niche practice. One monitoring company, Gaggle, says it works with about 1,500 school districts and tracks the online activity of roughly 6 million students. A U.S. Senate investigation found that more than 7,000 schools or districts used GoGuardian's monitoring products in 2021. In Vancouver, Washington, the district paid Gaggle $328,036 for three school years. The software scans what students search and write on school-issued devices, around the clock, and sends anything that matches its patterns to human reviewers and then to the school.

I am not arguing that schools should stop caring about safety or attendance. I am arguing that monitoring has grown faster than our habit of asking whether it works, who it hurts and who gets a say.

Who defines “risk,” and whose knowledge counts

Every monitoring tool carries a built-in definition of a problem. A keyword list decides which words signal danger. A behavior points system decides which actions count as disruptive. A vendor's model decides what a “concerning” search looks like.

Those definitions are usually written far from the classroom, by companies and administrators. The people with the most context are rarely consulted: the teacher who knows a student was quoting a song, the parent who knows why a child is late, the student who knows what a phrase means in their community.

Look at what that produces in practice. When the Associated Press and the Seattle Times examined Vancouver's records, they found that from October 2023 to October 2024 nearly 2,200 students, about 10% of the district, had been the subject of an alert. The system flagged more than 1,000 documents for suicide and nearly 800 for threats of violence, and many were false alarms. At one arts high school, about 1 in 4 students had written something that triggered an alert. One student was flagged for a short story with mildly violent imagery. That is what happens when a keyword list, not a teacher, decides what creative writing means.

When the record speaks louder than the people who know the student, the record wins. Foucault's point was that this is a form of power, even when everyone involved means well.

Who benefits, and who bears the costs

The benefits of monitoring are easy to name. Administrators get reports they can show a board. Vendors get contracts. Districts get a sense of having done something about safety.

The costs are harder to see, and they fall unevenly:

• Students who rely on school devices. A student with a personal laptop can search privately at home. A student whose only computer is school-issued may be monitored there too.
• Students already watched more closely. In a 2022 national survey by the Center for Democracy and Technology, 48% of Black students and 55% of Hispanic students said they or someone they knew had gotten in trouble because of monitoring, compared with 41% of white students.
• Students who need help. In the same survey, 29% of LGBTQ+ students said they or someone they knew had been outed by monitoring without consent. A teenager researching depression, sexuality or a family crisis may stop searching if they know it will be flagged.
• Students who were never asked. In the Vancouver records, at least six students were potentially outed to school officials after writing about being gay, transgender or struggling with gender dysphoria. In Durham, North Carolina, an advocate said a self-harm alert led to a student being outed to an unsupportive family, and the school board voted to stop using Gaggle in 2023. One Vancouver student, after learning about the monitoring, stopped searching personal health questions: “I was too scared to be curious.”
• Students whose secrets leak. The data collected to protect students can expose them. Through a single public records request, reporters received almost 3,500 sensitive, unredacted student documents from Vancouver's monitoring system, with no password or firewall in the way. Gaggle changed its system afterward. The lesson is simple: every record kept is a record that can be lost.
• Everyone's trust. Students who feel scored all day learn to perform for the system rather than to ask honest questions.

The same survey found that among teachers whose schools use monitoring software, 78% said students had been flagged for discipline and 44% said a student had been contacted by law enforcement because of an alert. A flag does not stay in the moment. It goes into a file, and that file can follow a student long after the situation has passed.

What the evidence and affected voices say

The public debate has already moved in places. In 2020 the Lockport school district in New York switched on a $1.4 million facial recognition system. Two parents, represented by the New York Civil Liberties Union, sued, and the state paused the technology. In 2023 a state report found the risks in schools “may outweigh the benefits,” including higher false-positive rates for people of color and transgender people, and warned it “may only offer the appearance of safer schools.” That September, New York's Education Commissioner banned facial recognition in schools.

Online monitoring has not faced the same test. No independent research shows that this software measurably lowers student suicide or violence. A 2023 RAND study found only “scant evidence” either way and concluded that “no research to date has comprehensively examined how these programs affect youth suicide prevention.” The Vancouver officials the reporters spoke to said they do not track whether the tool works, though they believe it has saved lives.

New York's ban happened because people asked a simple question: does this make students safer, and at what price? Most communities have not asked it yet. Here is what I think any district should be able to answer in public:

1. Which monitoring tools are in use, what they collect, and how long the data is kept.
2. How many flags or automatic penalties were issued last year, and how many led to a real intervention.
3. How those flags and penalties break down by student group.
4. Whether a student or family can see what was recorded about them, and how to correct it.

Numbers are only half of it. The other half is the people on the receiving end. Students, parents and teachers who agree to share their experience can say what a flag felt like, what it changed and whether anyone explained it. Those stories belong next to the vendor's marketing claims.

One reform: a student data bill of rights

I would start with one local policy, adopted by the school board and written with students, families and teachers at the table. It would rest on four commitments:

• Transparency. Publish every monitoring tool in use, what it collects and why, in plain language, every year.
• Collect less. No monitoring outside school hours on school devices without a specific, documented reason. No facial recognition.
• A human in the loop. No automatic penalty or referral based on a flag alone. A person who knows the student reviews it first.
• See it, challenge it, let it expire. Students and families can view their records, correct mistakes and appeal decisions. Minor discipline and risk notes expire after a set period instead of following a student forever.

This does not mean pulling support away from students who need it. A counselor can help a struggling student without turning every conversation into a permanent entry. Support should come with clear rules about who can see what, and the default should be the least sensitive record that still lets help happen.

Part of this already exists on paper. The federal student privacy law, FERPA, gives parents, and students once they turn 18, the right to see a student's education records within 45 days of asking, to request changes to anything inaccurate or misleading, and to a hearing if the school refuses. Few families know this, and it was written before a laptop could generate a report at 11 p.m. A local policy would make those rights easy to use and make clear that monitoring flags and risk notes count as records too.

How to know if it works. Give it one school year, then report back publicly on:

• The number of flags and automatic penalties, compared with the year before.
• How many flags led to real help, not just a record.
• Whether gaps between student groups narrowed.
• How many records were viewed, corrected or expired.
• What students and families say in a short survey about trust and fairness.

What I might be missing

The strongest case for monitoring is that it sometimes catches a student in real danger, such as someone searching for ways to hurt themselves, and gets them help in time. The alerts are not all noise. One Oklahoma district received close to 1,000 alerts in the 2023–24 school year, 281 of them about suicide. Among the screenshots reporters reviewed was a student asking, “Why does my boyfriend hit me?” Behind some of those flags are real children in real trouble, and I take that seriously. That is why this proposal keeps human review and support instead of banning monitoring outright. If a district's own numbers showed that its flags regularly led to real help, with little harm and no unequal impact, I would rethink how far these limits should go.

The real lesson

Schools teach more than their subjects. When students are watched and scored all day, they learn that the record matters more than they do. When they can see their records, question them and help write the rules, they learn something better: that institutions can answer to the people they serve.

If your school uses monitoring software, start by asking the four questions above at your next board meeting. The answers will tell you a lot.

Where the facts come from

Sources

• Center for Democracy and Technology, “Students and Teachers Warn That Schools’ Use of Monitoring Software Is Harming Students” (August 2022), the survey figures on discipline, law enforcement contact, outing, and differences by race.
• Associated Press and The Seattle Times, “Schools use AI to monitor kids, hoping to prevent violence. Our investigation found security risks” (March 2025), the Gaggle and GoGuardian figures, the Vancouver alerts and records leak, the Durham and Oklahoma examples, student quotes, and the RAND finding.
• WSKG / Associated Press, “New York bans facial recognition in schools after report finds risks outweigh potential benefits” (September 2023), the Lockport system, the NYCLU lawsuit, and the state report's findings.
• New York State Education Department, determination on biometric identifying technology in schools (September 27, 2023), the ban on facial recognition in schools.
• National Center for Education Statistics, model FERPA notice for school districts, the rights to inspect records within 45 days, request amendments, and get a hearing.
• Stanford Encyclopedia of Philosophy, “Michel Foucault”, an overview of Discipline and Punish: observation, normalizing judgment, the examination, and the Panopticon.