Does voice typing count as using AI?

It depends on the definition the rule uses. Technically, voice typing is AI, because the recognizer is a trained model. Whether your school, employer or exam board counts it as "AI use" depends on how its policy defines AI, and some policies do not use the broad definition. This page explains the common cases and what to do when a rule is unclear.

The broad definition covers speech recognition

The EU AI Act defines an AI system as a machine-based system that, for explicit or implicit objectives, infers from the input it receives how to generate outputs such as predictions, content, recommendations or decisions ¹. A speech recognizer takes audio as input and infers the most likely words, so under that wording it fits. A policy that copies this definition covers voice typing. The definition does not say whether a given product is covered, because that depends on the rest of the regulation, so treat it as a starting point for a policy rather than a verdict on one app.

Some rules sort tools by what they do

Some testing and school rules name tools by function, not technology. Colorado requires schools that use speech-to-text on state tests to disclose the name of the application, technology or program that students will use ². That is one state's spring 2025 form, so later years and other states may differ.

The ACT's list of allowable accommodations includes speech-to-text software, with conditions. It applies only if the examinee is registered for the writing test, unless dictation is used for the full test, and it is given in an individual room ³. That rule is about access, not about whether the tool is AI.

In England, the Department for Education's consultation judged that free-to-access speech-to-text tools are easy to get, simple to set up and support basic transcription, which makes them suitable for students with general dictation needs ⁴. Its March 2026 consultation proposed that, other than in exceptional circumstances, it would fund speech-to-text software only for students who, because of their disability, rely on speech recognition as their primary way of writing and controlling a computer ⁵. That is a proposal, not a rule in force.

A rule about generative AI may not reach a transcriber

A policy written against generative AI targets the rewrite step. A plain transcript is not new text written by a model, so the rule may not apply to it. A rewrite such as fluid dictation's correction of grammar and filler words is different. Microsoft describes that feature as powered by on-device small language models ⁶, which puts it on the generative side of the line.

Data rules are a separate question. Apple says Mac general dictation is processed on your device and not sent to Siri servers ⁷. A policy can allow a tool on data grounds and still require disclosure on AI grounds, so check both.

How to answer for your own policy

1. Read the definition. If it says "AI system", voice typing is in scope. If it says "generative AI" or names a product, check whether the rewrite step is on.
2. Check the tool's setting. Transcription and rewriting are often different switches in the same app.
3. Ask the owner of the rule. If the wording is unclear, ask the school, employer or exam board in writing which mechanism it means.
4. Keep a record. If a policy asks for disclosure, note the tool, the setting and whether a rewrite was on.

A clear policy names the mechanism it cares about, for example "transcription without generated text" or "generative rewriting of submitted work". Naming the mechanism avoids arguing about whether a tool is "AI" in general. This page explains the sources and is not legal advice on any one rule.

For the technical background, read the overview of whether speech to text is AI. The sibling questions go product by product: Does Microsoft dictation use AI?, Does Google voice typing use AI? and Is speech to text generative AI?. The guide to speech to text without AI explains how to keep dictation free of a rewrite step, and the overview of voice and AI lists the rest.