[ COMPARE ]

AI monitoring in web filters, and an approved AI

Web filters now sort AI apps and log student prompts on school devices. What that monitoring does, what it cannot do, and how a filter and an approved AI work together.

Plan a pilot Watch the demo

Facts as of Oct 18, 2026Last reviewed Oct 19, 20263 min read

[ SUMMARY ]

The short version

Web filters have become schools' AI approval desk. Lightspeed, GoGuardian and Securly now sort AI apps, log student prompts on school devices and flag concerning messages. That is monitoring. It does not give students an approved place to get help, or set how much help the AI gives. A school that adopts Nexus keeps its filter: allow Nexus, keep consumer AI blocked.

What the filter vendors offer

Facts as of Oct 18, 2026, from each vendor's pages:

Lightspeed Systems. An AI governance dashboard sorts AI apps into approved, restricted and blocked by grade or campus, detects new AI tools, logs every prompt students send to AI tools in real time, and alerts on violence and self-harm. Its 2026 app report found 406 new AI apps in district use. Its Prompt Capture feature "does not block or modify AI interactions."
GoGuardian. AI governance through Admin, Discover (AI adoption and approval status) and Beacon, which flags concerning prompts. Prompt visibility and prompt blocking are slated for 2026-27.
Securly. AI Transparency lets admins read student ChatGPT and Gemini prompts and replies, shows teachers prompts typed in class, offers parents a view if the district allows it, deflects off-policy prompts and raises self-harm alerts. The same page describes Securly AI Chat, a managed student chat where the district picks the model.

Monitoring is not the same as learning rules

A filter can see a prompt or stop one. It cannot make an AI give a hint instead of an answer, or keep a student working through the steps. Those rules have to live inside the AI the student uses.

The research on answer-giving AI is why that matters. In a field experiment with about 1,000 high-school students, answer-giving AI raised practice scores but cut unaided exam scores by 17%. A hint-first version of the same AI removed that harm.

Side by side

Item	Filter-based AI governance	Nexus
Main job	Sort, allow, block and log AI use on school devices	Give students an approved AI for schoolwork
Learning rules	None inside the AI; filters can block or deflect prompts	A class learning mode students cannot change
What staff see	Prompts on school devices (varies by vendor and setting)	Class chats, with notice to students; every read logged
Personal use	Can capture anything typed on a managed device	Personal chats are not shown to teachers

Filter-based AI governance and Nexus, as of Oct 18, 2026

What filters mean for student privacy in Nexus

Personal chats in Nexus are not shown to teachers. But monitoring software on a school device works outside Nexus and may capture anything a student types, in any app. Tell students and families both facts plainly. Research on school monitoring shows why: about half of monitored high schoolers agreed they hold back their true thoughts or ideas online because they may be monitored (CDT, via Education Week, 2022).

How the two work together

Ask your filter vendor how Nexus is classified, then allow it while consumer AI stays blocked. We send your IT team the exact list of addresses to allow, for GoGuardian, Lightspeed, Securly, Linewize or another filter. Test the allow-list on a school device on school Wi-Fi before students start. The steps are on Nexus and your web filter.

[ FAQ ]

Questions and answers

Can web filters see student AI prompts?

Some can, on school devices. Lightspeed logs prompts students send to AI tools in real time, Securly lets admins read student ChatGPT and Gemini prompts and replies, and GoGuardian has prompt visibility slated for 2026-27 (vendor pages, read Oct 18, 2026).

Do we still need a filter if we adopt Nexus?

Yes, for most schools. The filter keeps consumer AI blocked on school devices; Nexus gives students an approved AI with class rules. Allow Nexus in the filter.

Does monitoring AI prompts improve learning?

No study we know of shows that. Monitoring can catch some risks, but learning rules have to live inside the AI students use.

Will the filter see students' personal Nexus chats?

Possibly, on a school device. Personal chats are not shown to teachers in Nexus, but monitoring software on a school device works outside Nexus and may capture anything a student types.

[ SOURCES ]