AI behind the scenes.
Reading in the foreground.

How artificial intelligence appears in our product, how schools can use LitLab with or without student screens, and how teachers stay in control.

✓No chatbot✓No open-ended prompts✓Teacher-reviewed scores✓~15 min, twice weekly✓No student screen required
Assigned decodable · short a
Sam sat on the mat and had a nap.
● student reading aloud
Student reads aloud
AI listens and analyzes
Student practices more

LitLab supports authentic reading practice. Teachers stay in control.

Effective phonics practice requires students to read connected text aligned to concepts they have been explicitly taught. Teachers need enough appropriate texts to provide that practice, and they need opportunities to hear students read so they can identify where closer attention may be needed.

Educator tools

Curriculum-aligned decodables, with or without student screens

AI allows LitLab to provide more varied and personalized texts at a specific point in a phonics sequence than a teacher or publisher could feasibly create manually for every skill, group, and student.

Teachers choose from 1,500+ educator-reviewed decodables or generate and preview a new text aligned to the phonics skill they have taught. Students access texts through print, projection, or a device.

Educator-reviewed libraryTeachers previewNo student screen required
Student practice

Short, purposeful screen time for oral reading practice

When the Read Aloud feature is used, LitLab provides teachers with preliminary information about words correct per minute, accuracy, and possible miscues. This allows them to collect and review oral reading information from more students than they could hear individually during the same period of instruction.

The analysis is not a grade, a placement decision, or the sole measure of a student’s reading. Teachers can listen to the recording, confirm or correct the analysis, and interpret it alongside classroom observation, curriculum assessments, and other student work.

Oral reading~15 min, twice weeklyTeacher can verifyNo gamificationSecure
The short version

What students experience with Read Aloud.

Student screen time is optional and bounded. Printed and projected use requires no student screen time. When Read Aloud is used, LitLab recommends approximately 15 minutes twice weekly. LitLab does not use coins, streaks, or rewards designed to keep students online longer.

Short adecodable
sh · chdecodable
Teacher control

The teacher picks the skill and the text.

The assignment comes from the teacher’s scope and sequence. Students are not routed into a separate adaptive curriculum that drifts away from classroom instruction.

✓The teacher determines what the child practices.
The AI inventory

Every place AI appears.

Filter by who interacts with the feature and click on a card to see how it works, what the student experiences, and what safeguards or human review always apply.

Most AI-supported capabilities are teacher tools or operate behind the scenes. Student-facing experiences are bounded: schools can disable Story Creator, and teachers decide when to assign Read Aloud.

7 of 7 capabilities shown
MethodLanguage-model generation constrained by a deterministic scope-and-sequence engine. A grapheme-level checker rejects and regenerates text below the decodability threshold; illustrations use an image-generation model.
Student interactionNone — this is a teacher tool.
Human oversightAutomated decodability check; the teacher can preview and edit the story before assigning.
MethodAutomatic speech recognition, with two models cross-checked to produce a confidence score.
Student interactionThe child reads a book aloud. There is nothing to type.
Human oversightA teacher can confirm or correct every score.
MethodThe same constrained decodable-generation pipeline.
Student interactionPicture-card choices only. No text field, prompt, or chat. Limited to three per day.
Human oversightTeachers and district leaders can turn it off; all output is visible to the teacher.
MethodGenerated through the same constrained pipeline, then reviewed and edited by trained educators.
Student interactionNone.
Human oversightEvery text is reviewed by a trained educator before publication.
MethodRecommendation model.
Student interactionNone — teacher-facing only.
Human oversightMakes recommendations; the teacher decides what to do.
MethodDeterministic mapping on oral reading analysis results; teacher-verified results are tracked separately from unverified ones.
Student interactionNone.
Human oversightTeacher can optionally verify data for recordings.
MethodAlgorithmic filtering on inputs and outputs, plus model-based filtering on outputs.
Student interactionNone.
Human oversightAutomated.
Hard boundaries

What a student never does in LitLab.

These guardrails are built into the product. They are not usage guidelines that can be ignored or overridden.

×

Students do not type anything into a model

No text input, prompt box, chat window, or free-form field in the student experience.

×

Students do not talk to a chatbot or tutor

Nothing converses with a child, answers their questions, or responds conversationally to what they say.

×

Students do not reach general-purpose AI

Students see only material generated inside the phonics constraints their teacher set.

×

Students do not receive an AI score directly

AI-generated reading scores go to the teacher, who can review and correct them before they are acted on.

Design principle

Teachers make the instructional decisions.

LitLab gives teachers access to reviewed, curriculum-aligned texts and, when Read Aloud is used, a recording of each student’s oral reading with preliminary analysis the teacher can confirm or correct. Teachers decide what students practice, how to interpret the information alongside other evidence, and what instructional response should follow.

Privacy & data

How we handle recorded student voice data.

When a student reads aloud, the recording is transmitted for analysis and to the teacher, and stored so the teacher can listen to it. Here is the path and the guardrails around it.

🎙️
Child readsassigned decodable
🔒
Encrypted transitrecording sent securely
◎
Speech analysisreading metrics + miscues
🧑‍🏫
Teacher reviewlisten, confirm, correct
✓

Not sold. Ever. Not aggregated, not de-identified, not at all.

✓

Not used to train third-party AI models. Provider agreements prohibit training on what LitLab sends.

✓

Not used for advertising or marketing. Never shared outside the school for someone else’s purposes.

✓

Stored in the United States and encrypted. Designed for FERPA, COPPA, Illinois SOPPA, and New York Education Law 2-d requirements.