WCAG 3.0 in Late 2026: Bronze, Silver and Gold Are Gone. Here's the Conformance Model That Replaced Them

If you searched for WCAG 3.0 this month, most of what you found is wrong.

Search results are full of 2026 explainers describing a conformance model with Bronze, Silver and Gold tiers, a 0-5 scoring system, and roughly 174 outcome-based requirements. That description was a reasonable summary of an earlier draft. It is not what the W3C published in September 2026.

The WCAG 3 Introduction page, which the W3C Web Accessibility Initiative maintains as the canonical starting point for the standard, was last updated on 10 September 2026. It describes something different: a single conformance level, with tags providing reporting tiers above and below it. Bronze, Silver and Gold are not in the current proposal.

Before the rest of this post, the part that matters most: none of this changes your legal obligations today. The European Accessibility Act's technical floor runs through EN 301 549, which references WCAG 2.1 AA. A W3C Working Draft has no legal force anywhere in the EU, and will not acquire any for years. If you are mid-remediation, keep going.

What the W3C actually published in September 2026

Two documents were updated.

The Explainer for WCAG 3 provides background and describes the structure of the draft. The W3C's own advice is to read the Explainer before reading the standard itself, which is unusually direct guidance and worth following. Its conformance section was revised in the September update.

The WCAG 3 Working Draft was updated to include guidelines, requirements and assertions that have progressed to "Developing" status. Anything still at "Exploratory" status was deliberately held back and lives only in the Editor's Draft. That distinction is doing a lot of work, and we come back to it below.

One naming point that still confuses people in procurement documents. WCAG 2 stands for Web Content Accessibility Guidelines. WCAG 3 stands for W3C Accessibility Guidelines. The acronym was kept because it is universally recognised; the expansion changed because the scope now reaches beyond web content to apps, tools, publishing and emerging technologies.

The conformance model, as it stands today

The current proposal, in the W3C's own words, "sets a single conformance level and uses tags to provide reporting tiers below and above that level."

Read that carefully, because it inverts the mental model most people have absorbed. There is one bar. You either meet it or you do not, in the same binary sense as WCAG 2 Level AA. What the tags add is reporting granularity: a structured way to describe how far below the bar you currently sit, or how far above it you have chosen to go. The tiers are a communication device layered on a single conformance decision, not a menu of conformance levels you pick between.

This is not settled. The Explainer carries a live alternative under the heading "Alternative approach - Scoring", which uses the same tags to generate a numeric score above and below conformance. The Accessibility Guidelines Working Group is openly asking which is better, and has published three review questions:

• Does the single-level-plus-tags approach improve on the conformance approach in previous drafts?
• What are the pros and cons of each approach?
• In your role, how would you use reporting tiers leading to conformance, or above conformance?

A working group still asking "which of these two fundamentally different models should we use" is not a working group approaching a stable answer. The practical implication is blunt: any organisation building a compliance programme, a vendor scorecard, or a product roadmap around a specific WCAG 3 tier scheme today is building on ground that has already shifted once and is being actively invited to shift again.

Assertions: the part EAA teams should actually care about

Buried in the September update is the change with the longest shadow for regulated organisations. Assertions have progressed to Developing status alongside requirements.

An assertion is not something a scanner can find. It is a governance claim: a statement that the organisation did a thing, documented in a way that can be inspected. Ran usability testing with disabled participants. Trained the front-end team. Maintained a testing protocol across releases. Recorded the decision and the rationale when an exception was taken.

This matters because it extends accessibility conformance out of the rendered DOM and into organisational policy and documented process. Automated tooling cannot generate an assertion. Neither can a remediation sprint.

If that sounds familiar, it should. It is the same shape as obligations EAA-covered organisations already carry. An accessibility statement is a public, documented claim about how a service meets requirements and what it does when it does not. The Annex IV technical file behind a CE-marked product is a documented evidence trail that a market surveillance authority can demand. The EAA has been pulling private-sector accessibility towards documented process for a while; the WCAG 3 draft is pulling the technical standard in the same direction.

The useful conclusion is not "prepare for assertions". It is that the record-keeping WCAG 3 will eventually want is record-keeping EU market surveillance authorities already ask for. Build it now, for the regulator you have, and the standard you get in several years will find you ready.

Read the section statuses before you read the draft

The single most useful thing a practitioner can learn about WCAG 3 is not any requirement. It is the status ladder attached to each normative section:

1. Placeholder - a heading with intent behind it, nothing more.
2. Exploratory - an idea under active investigation. May vanish entirely.
3. Developing - has survived initial scrutiny and is being built out.
4. Refining - substantially settled; detail work remains.
5. Mature - stable enough to rely on.

Almost every misleading WCAG 3 article shares one root cause: it quoted an Exploratory section as though it were settled policy. The Bronze/Silver/Gold narrative spread this way.

Some sections additionally carry a "Needs additional research" flag, and the W3C is explicitly asking for help identifying or conducting studies to support or refute them. A requirement marked that way is a question, not an answer.

Practical rule for anyone monitoring the standard: ignore everything below Developing. Skim Developing. Pay attention to Refining and Mature, of which there is currently very little.

The real timeline, and the EN 301 549 pipeline behind it

The W3C's own language is unambiguous. WCAG 3 "is currently an incomplete draft that will change" and "is not expected to be a completed W3C standard for a few more years."

Two further statements from the same page deserve to be quoted in every internal briefing on this topic. First, WCAG 3 will not supersede WCAG 2. Second, WCAG 2 will not be deprecated for at least several years after WCAG 3 is finalised. The two standards will coexist, and the transition will be gradual and signposted. The Accessibility Guidelines Working Group maintains a projected schedule in its GitHub wiki for anyone who wants to track it.

Third-party commentators commonly project a Candidate Recommendation around 2027 and a Recommendation no earlier than 2028. Those are outside projections, not W3C commitments, and should be labelled as such whenever they appear in a board paper.

Now add the European layer, which almost nobody models correctly.

WCAG 3 becoming a W3C Recommendation would not, by itself, change anything under the EAA. The EAA does not cite WCAG directly. It relies on harmonised European standards, principally EN 301 549, and a harmonised standard only confers presumption of conformity once it has been developed under a Commission standardisation request and its reference has been cited in the Official Journal of the European Union.

So the chain runs: W3C finalises WCAG 3, then the European Commission issues or amends a standardisation request, then CEN, CENELEC and ETSI revise EN 301 549 to reference it, then the revised standard is assessed and its reference published in the Official Journal, and only then does building to WCAG 3 buy you presumption of conformity under the EAA. Each of those links has historically taken years, and they run in sequence.

We are not going to invent a date for the end of that chain. The mechanism is the point. Anyone telling you EU law will require WCAG 3 by a specific year is describing a process whose timings are not yet knowable.

APCA and contrast: the honest answer

APCA, the Advanced Perceptual Contrast Algorithm, generates persistent search traffic and persistent confusion.

Here is the accurate position. APCA-based approaches to contrast have been explored in WCAG 3 work. Exploration is not adoption. APCA is not a legal requirement in any jurisdiction, is not referenced by EN 301 549, and does not appear in any enforcement standard an EAA market surveillance authority would apply.

For everything that touches your EAA exposure, you test against WCAG 2.x contrast: 4.5:1 for normal text, 3:1 for large text and for user interface components and graphical objects. Colour contrast remains the most common WCAG failure on the web by a wide margin, and it is the failure category where cheap fixes produce the largest measurable improvement. Spend the effort there, not on speculative contrast maths.

Nothing stops a design team using APCA as an internal quality signal on top of the legal minimum. Just never let it replace the ratio you will actually be measured against.

What to do now

Keep building to WCAG 2.2 AA. The W3C's advice could not be plainer: "The best way to prepare for WCAG 3 in the future, is to meet WCAG 2 Level A and AA success criteria now." WCAG 2.2 AA subsumes 2.1 AA, which is the EAA floor via EN 301 549, so this single target satisfies today's law and tomorrow's transition simultaneously.

Start keeping the evidence assertions will want. Testing protocols that produce comparable results release to release. Training records. Decision logs for every exception taken and every disproportionate-burden argument considered. Audit reports with dates and scope. None of this is speculative investment; a market surveillance authority can ask for it this quarter.

Monitor statuses, not headlines. Subscribe to W3C WAI news directly rather than reading it secondhand. When a WCAG 3 article crosses your desk, check the section status of whatever it is describing before you circulate it.

Participate if you have standing. The working group is actively soliciting views on the conformance model, particularly from people who would have to implement and report against it. Comments go to the WCAG 3 GitHub repository, or to public-agwg-comments@w3.org if GitHub is not workable. Compliance leads at EU-facing businesses are precisely the constituency the review questions are aimed at, and that constituency is under-represented.

Treat WCAG 3 claims as a procurement red flag. No product can be "WCAG 3.0 ready" or "Gold certified" against a standard whose conformance model changed this month and may change again. A vendor making that claim is either not tracking the standard or is willing to say things that are not true, and both should affect your scoring. File it next to overlay widgets in your supplier due diligence.

Do not rewrite your design system around a draft. Every hour spent restructuring components for a hypothetical conformance model is an hour not spent on the contrast failures, keyboard traps and unlabelled form controls that are in scope for enforcement right now.

Frequently asked questions

When will WCAG 3.0 be released? The W3C says it is not expected to be a completed standard for a few more years, and declines to give a date. Third-party projections cluster around a Candidate Recommendation in 2027 and a Recommendation no earlier than 2028, but these are projections rather than commitments.

Are Bronze, Silver and Gold still a thing? Not in the current proposal. The September 2026 draft sets a single conformance level and uses tags for reporting tiers above and below it. An alternative scoring model remains under consideration. Any article describing Bronze, Silver and Gold as the WCAG 3 conformance model is describing an earlier draft.

Does WCAG 3.0 affect my EAA obligations? No. The EAA operates through EN 301 549, which references WCAG 2.1 AA. A W3C Working Draft is not a harmonised standard and confers nothing. Your obligations will only change when a revised EN 301 549 referencing WCAG 3 is cited in the Official Journal, which is several procedural steps and several years away at minimum.

Should I wait for WCAG 3.0 before remediating? No, and the question usually signals a desire to defer spending. Enforcement is active now across EU member states. WCAG 2 will remain valid for years after WCAG 3 is finalised, meeting WCAG 2 AA is the W3C's stated best preparation for WCAG 3, and waiting simply accumulates both legal exposure and technical debt.

Will WCAG 3.0 replace WCAG 2.2? No. The W3C states explicitly that WCAG 3 will not supersede WCAG 2, and that WCAG 2 will not be deprecated for at least several years after WCAG 3 is finalised. Expect a long period in which both are current and guidance on transitioning between them is published well in advance.