ADA video captions: two titles, two kinds of obligation

The Americans with Disabilities Act does not contain a sentence that says "caption every video". The obligation comes from two places. Title II covers state and local governments, and since 2024 it has a dedicated rule for web content and mobile apps with a named technical standard. Title III covers businesses and nonprofits that serve the public, and for them the Justice Department has not adopted a web-specific technical standard; the general duty to communicate effectively with people who have hearing disabilities applies instead.
Which title you fall under changes almost everything: the standard you measure against, the date, and how much room you have to decide what an accessible video looks like. This article explains both in plain language, gives a captioning checklist, and points to the official sources. It is not legal advice; for a decision about your organisation, talk to counsel or the ADA Information Line run by the Justice Department.
Title II: the 2024 web rule for state and local governments

On April 24, 2024, the Federal Register published the Justice Department's final rule updating its Title II regulations. The rule makes the Web Content Accessibility Guidelines (WCAG) version 2.1, Level AA, the technical standard for web content and mobile apps that state and local governments provide or make available, including content a contractor or vendor publishes for them. The ADA.gov fact sheet on the rule names captions for videos as an example of what WCAG requires, and the rule's definition of web content includes sound and videos.
The compliance dates were then moved. On April 20, 2026, the Federal Register published an interim final rule that extended them:
- 50,000 or more persons
- April 26, 2027
- 0 to 49,999 persons
- April 26, 2028
- Special district governments
- April 26, 2028
Any page that still gives April 24, 2026 and April 26, 2027 as the deadlines is quoting the original 2024 schedule, which the interim rule replaced. ADA.gov notes that a school district is not a special district government: a city or county school district uses the population of that city or county, and an independent school district uses the most recent Small Area Income and Poverty Estimates figure. A city library or police department uses the population of the city that runs it.
Who Title II covers

Title II applies to every state and local government agency and department, plus special purpose districts, Amtrak and other commuter authorities. The fact sheet's examples include public schools, community colleges and public universities, courts, elections offices, police departments, public hospitals and clinics, parks and recreation programs, public libraries and public transit agencies. If a city records council meetings, a county posts a public health explainer, or a state university publishes lectures, that video is web content under the rule.
The rule has limited exceptions that matter for video archives. Content created before the compliance date, kept only for reference, research or recordkeeping, stored in a dedicated archive area and not changed since archiving does not have to meet WCAG 2.1 AA, but only if all four conditions hold. Social media posts made before the compliance date are also excepted. ADA.gov is clear that the exceptions do not remove the duty of effective communication: its own example is a person with a hearing disability asking for an archived city video, where the city could caption it and share the captioned file.
Title III: businesses and nonprofits that serve the public

Title III covers public accommodations: shops, hotels, private schools, clinics, gyms, entertainment venues and many other businesses and nonprofits. ADA.gov's effective communication guidance says these entities must provide auxiliary aids and services when needed to communicate effectively with people who have hearing, vision or speech disabilities, and its list of aids and services explicitly includes open captioning, closed captioning, real-time captioning and closed caption decoders.
What Title III does not have is a regulation that names WCAG for websites. Courts decide disputes case by case, and many organisations treat WCAG 2.1 AA as the practical yardstick because it is the standard the Justice Department chose for Title II. Undue burden is judged against the size and resources of the business, so expectations of a large retailer differ from those of a small studio. If your videos are how customers learn about, buy or use your service, uncaptioned video is a gap worth closing whatever the legal analysis says.
What a caption has to contain

Both titles point back to the same idea of a usable caption. WCAG success criterion 1.2.2 asks for captions on all prerecorded audio in synchronized media, and W3C's explanation says captions carry dialogue, who is speaking and meaningful non-speech sounds. Live video falls under 1.2.4 at Level AA, and prerecorded video also needs audio description under 1.2.5 when visuals carry information the soundtrack does not. Rather than repeat each criterion here, WCAG caption requirements walks through them one by one, and how to write audio description covers the visual side.
Accuracy is part of the requirement, not a nice extra. In its 2022 complaint against the University of California, Berkeley, the Justice Department alleged that many of the university's public videos had no captions or relied on automatic captions that were inaccurate, lacked punctuation and were difficult or impossible to understand. Captions that exist but cannot be followed do not deliver effective communication.
A practical checklist for captioning video

- Find out which title applies. Public body or public school: Title II and its date. Business or nonprofit serving the public: Title III and effective communication.
- Inventory your video by place: website pages, embedded players, course platforms, mobile apps and social accounts. Note which items were posted before your compliance date and whether they sit in a true archive.
- Prioritise video people need right now: meeting recordings, service explainers, emergency and safety information, enrolment and benefit instructions.
- Produce a first draft of captions from the audio, then have a person correct names, numbers, technical terms and punctuation against the recording.
- Add speaker identification where the speaker is not obvious on screen, and meaningful sounds such as [applause], [alarm sounds] or [music stops]. Captioning sound effects and music shows how to word them.
- Check timing and reading: no overlaps, no flashes too short to read, two lines at most.
- Publish captions as a closed track the viewer can turn on, or burn them in where the platform has no caption support, and confirm they play in the actual player.
- Plan for live events separately. Prerecorded workflows do not cover a livestreamed meeting; how live captioning works explains the options.
- Write down how people can request an accessible version of anything you have not captioned yet, and answer those requests quickly.
A county of 120,000 people streams its board meetings and posts the recordings the next day. Its date is April 26, 2027. Staff caption each posted recording, add speaker names because the camera stays wide, and arrange live captions for the stream itself.
A town of 8,000 people has a two-minute video on booking picnic shelters. Its date is April 26, 2028, but the video explains a current service, so it goes to the top of the list rather than waiting.
A clinic posts a video on aftercare for implants. It falls under Title III, so no compliance date applies, but a patient who is deaf needs the instructions as much as anyone; the clinic adds closed captions and a text version.
Sponsored · mydubly
AI video dubbing deal
Dub any video for just 12¢ a minute
Was 240 credits/min — now only 120. Cloned voices, timed to your picture, at half the list price.
Where AI captions fall short
Speech recognition gives you a fast, cheap first draft, and that is its honest role. It does not know who is speaking, it does not describe sounds, and it can get a council member's name, a street, a dosage or a statute number wrong with full confidence. Crosstalk in meetings, poor room microphones and heavy accents all make errors more likely. A draft that nobody has checked is the kind of automatic caption the Berkeley complaint describes. Budget for human review on every file; how to proofread an AI transcript gives a method that works for caption files too.
Where mydubly fits
The subtitle generator turns a recording's speech into SRT and VTT caption files in 100 languages for 5 credits per minute, with a minimum of 5 credits per file, so a 45-minute meeting recording costs 45 credits (4.5¢). The output is a draft: it contains the recognised words with timing, but no speaker labels or sound descriptions, and a person must review it for accuracy before you rely on it. The free subtitle editor lets you correct lines, add speaker names and sound tags, fix timing with warnings for fast reading speed, long lines, very short cues and overlaps, then export SRT or VTT or render an MP4 with burned-in captions. mydubly works on recorded files only, does not produce live captions or audio description, and does not make anyone compliant; that depends on your content, your process and your legal obligations. For school-specific planning, captions for educational videos goes further.
Sources used
- ADA.gov fact sheet on the Title II web and mobile app rule, including the April 2026 interim final rule and the compliance date table.
- ADA.gov, ADA Requirements: Effective Communication, for Title III and the list of auxiliary aids and services.
- W3C, Understanding Success Criterion 1.2.2: Captions (Prerecorded).
- United States v. The Regents of the University of California, the Justice Department's 2022 complaint.
This article summarises public guidance and is not legal advice.
Sponsored · mydubly
240 free credits + 50% off
Try free, then buy minutes at half price
New accounts get 240 credits (enough for a 2-minute dub). When you top up, you keep these sale rates.
Generate subtitles from speech
Create SRT and VTT files from the spoken audio, in the original language or translated, ready to upload to any player. To fix, style or burn subtitles into an MP4, open the free subtitle editor.
Frequently asked questions
What are the current ADA Title II deadlines for accessible video?
Under the interim final rule published April 20, 2026, state and local governments with 50,000 or more people must meet WCAG 2.1 Level AA by April 26, 2027. Entities under 50,000 people and all special district governments have until April 26, 2028.
Does the ADA require captions on a private company's marketing videos?
Title III has no web-specific technical rule, but businesses that serve the public must communicate effectively with people who have hearing disabilities, and ADA.gov lists open and closed captioning among the aids that can achieve that. Whether a particular video must be captioned is a legal question for your situation, so get advice if it matters.
Are YouTube automatic captions enough for ADA purposes?
Not on their own. The Justice Department's 2022 complaint against UC Berkeley described inaccurate automatic captions without punctuation as difficult or impossible to understand. Treat automatic captions as a draft and have a person correct them.
Do old videos on a government website need captions by the deadline?
Only if no exception applies. Video created before the compliance date that is kept for reference in a dedicated archive area and has not been changed can be excepted, as can social media posts made before that date. A person who needs an archived video can still request an accessible version.
Does a livestreamed public meeting count under the Title II rule?
WCAG 2.1 Level AA includes captions for live audio in synchronized media, so live streams that a covered entity provides are in scope. Live captioning needs a real-time service; a workflow for recorded files does not cover it.
Can mydubly make my videos ADA compliant?
No tool can promise that. mydubly produces draft SRT and VTT captions from recorded speech and a free editor to correct them; meeting the ADA depends on accurate, complete captions, your other content and your obligations as a whole.
Sponsored · mydubly
Transcripts & subtitles deal