YouTube Will Label Undisclosed Brand Deals Itself: What the New Branded Content Policy Changes

Now YouTube ticks the sponsorship box for you

A brand sends the product, the creator films it, and the video goes up on a Thursday afternoon with the paid promotion box left unticked. Nothing happens. That has been the shape of the risk for years: an omission that nobody notices until a regulator does, months later, when the campaign is already over and the invoice is already paid.

YouTube published a new Help Center policy page that changes who closes that gap. It is titled "YouTube Branded Content Policies", it gathers rules that used to live in the paid product placement article, and it adopts branded content as the operative term across the whole subject. One sentence on it does most of the work: "If our systems detect your video includes branded content you have not disclosed, we may automatically apply a branded content label to your video and notify you." Two trade outlets reporting independently, PPC Land and Relevant Audience, date the announcement to September 3, 2026 and trace it to a post by a TeamYouTube community manager in the Help Center Community. Both Help Center pages quoted here were read directly. That community thread was not, so everything sourced to it is attributed to the outlets that quote it.

Two announcements landed the same day and some coverage stapled them together. This post keeps them apart, because YouTube did. The disclosure rewrite is one story. A limited Shopping Affiliate experiment about which local retailer serves a tagged product is a different one, and it explicitly leaves branded content alone. If you have seen the phrase "branded content sourcing" in the last week, it is a headline artefact, not a rule.

What YouTube published, and where to read it

The trade write-ups are useful for the date. The pages themselves are better for everything else, and both are readable without an account.

The new page states the rule in one sentence

The policy opens with a definition broad enough to catch most of what agencies actually do: "Branded content is any content on YouTube influenced by a brand partner in exchange for something of value, such as featuring or promoting a brand's products or services in exchange for payment, free products, or sponsorships". Gifted product is in there by name. So is a benefit that arrives later.

Scope is drawn just as wide. YouTube writes that "This policy applies to branded content wherever it appears, including videos, video descriptions, comments, live streams, Shorts, and any other YouTube product or feature." That closes the oldest workaround in creator marketing, which was to keep the sponsor out of the video and put it in the description instead.

And enforcement reaches the whole upload rather than the sponsored minute: "This policy applies to the entirety of your content, not just the portion where the branded content or sponsorship appears." One product mention at the eleven minute mark puts the entire video inside the policy.

The old page moved with it, which is how you know this is a migration

The article most creators have bookmarked now carries the title "Add branded content restrictions & disclosure labels". Its own history is the cleanest evidence there is: an archived capture of the same URL from April 1, 2026 is titled "Add paid product placements, sponsorships & endorsements", and the phrase branded content appears nowhere on that version. Same address, different vocabulary, five months apart. The live page opens in the new one: "If you feature branded content, sponsorships, endorsements, or other commercial relationships in your videos, you have to let YouTube know by selecting the paid promotion button in your video details." It defers to the new document in a line you can check yourself: "For more details, review YouTube's Branded Content Policy".

Even the checkbox copy changed. The Studio instructions now tell you to select "Yes, my video includes branded content". That is the tell that this is a terminology consolidation rather than a new prohibition, and it is observable on YouTube's own surface rather than inferred from reporting.

New, versus what was already true

Sorting these matters, because the parts that were already in force are the parts you may already be out of compliance with.

Item	Status	Where it is stated
Policy rewritten around "branded content"	New	Both Help Center pages, observable today
Refreshed viewer-facing disclosure label	New, described as rolling out	YouTube quoted by Social Media Today
Age and country restrictions at declaration	New	Both Help Center pages
Automatic labelling of undeclared brand deals	Announced, "in the coming months"	Community announcement, quoted by two outlets
The paid promotion declaration in Studio	Long-standing	The legacy article, rewritten not replaced
EU political paid promotion declaration	In force since October 10, 2025	The legacy article

The automated label, read slowly

This is the half worth being careful with, because the two published descriptions of the same mechanism do not say the same thing.

The word "may" is doing real work

The Help Center sentence is permissive rather than promissory. YouTube says it "may automatically apply a branded content label to your video and notify you". A policy page can carry a rule the systems behind it are not yet running, and here it does. The community announcement, as quoted by PPC Land and by Relevant Audience, describes YouTube as "rolling out internal automated detection systems in the coming months to help enforce our mandatory disclosure policies". So the correct present-tense statement is that the rule exists and the enforcement is scheduled without a date.

Relevant Audience is blunt about how little else was published: there is no launch date beyond "in the coming months", no country list, no penalty schedule, and no description of what the classifier actually reads. If anyone tells you the detection is live in your market this week, ask them where that is written down.

The scopes in the two documents do not match

Here is the detail most summaries have flattened, and PPC Land flagged the same gap. The Help Center page says "your video". The community announcement, as quoted by Social Media Today, says: "If our systems detect that your newly uploaded video includes branded content that you did not disclose, we may automatically apply the disclosure label on your behalf to ensure policy compliance and viewer transparency."

Where it is written	The scope it names	What that would mean
The Help Center policy page	"your video"	Anything on the channel, back catalogue included
The community announcement, as quoted by two outlets	"your newly uploaded video"	Only uploads after an unnamed switch-on date

Newly uploaded is a much smaller promise than your video. One reading covers everything going forward from an unnamed switch-on date. The other reads as though a back catalogue is in scope. Nothing published resolves it, so the safe planning assumption is the wider one, and the honest thing to say about the narrower one is that it appears only in the announcement. It is the same reading discipline that three platforms' AI disclosure rulebooks reward: the summary and the policy page rarely carry the same scope, and only one of them is enforceable.

The override is conditional, not an appeal route

There is a way back if the label lands on a video that never had a sponsor, and its wording deserves a second read: "If you feel the label was applied in error, you may have the option to certify that your video does not contain branded content and override the label."

May have the option is not a guaranteed process. There is no published description of how that certification is surfaced in Studio, no stated review window, and no false-positive rate. For an organic review of a product you bought yourself, that is the whole risk in one clause: a machine decides the video looks sponsored, the label appears, and your route back is conditional.

The age and country gates are the sleeper change

Automatic labelling is the headline. The restriction controls are the thing that will actually change how a campaign is planned, because they turn a brand deal's audience into a setting somebody has to choose.

Three limits, and they only appear once you declare

Tick the paid promotion box and YouTube says you get to set three things:

• Geographic locale where your content will be shown
• Overall minimum age to view your content
• Minimum age to view your content in specific geographic locales

The paid product placement article explains what they are for: "You can limit who gets to find your branded content based on their age or location. These manual controls help you meet local advertising laws, the requirements of your brand partnerships, and comply with YouTube's Community Guidelines and Google Ads policies."

Read the sequence rather than the list. The controls surface after the declaration, which means the compliance tooling is a reward for declaring rather than a burden. That is a design choice, and it is a smarter incentive than a penalty.

Restricted categories are where this bites first

YouTube names the categories that carry extra obligations: "Alcohol Financial Services Healthcare and Medicines Gambling Elections and Political Content". For those it puts a duty on the creator that most creator contracts do not currently mention: "you are responsible for confirming that your brand partners are certified by Google as necessary."

If you sell supplements, a financial product or anything drinkable, that sentence is a line item for your next creator brief. It is also the reason the geographic control exists rather than a blanket ban, since the rules differ by country and YouTube has pushed the choice down to the person uploading.

There is a hard edge too. The prohibited list, which no restriction control unlocks, reads: "Recreational drugs or paraphernalia Weapons or ammunition Hacking software Counterfeit products An academic essay-writing service". YouTube adds that the list is not complete, so read it as a direction of travel rather than a boundary.

The EU political rule was already here

One date on the legacy page predates all of this: "Starting on October 10, 2025, due to new European Union regulations on political advertising, Google will no longer allow political advertising in the EU". Creators have had to declare an EU political paid promotion since then, and content declared that way becomes unavailable to EU users. It is worth separating from the September update, because it is a live obligation rather than a coming one.

The affiliate test is a different announcement

Now the second story, kept in its own section on purpose. Two announcements, one day, and almost nothing in common:

	The disclosure update	The affiliate experiment
What it touches	Declared brand deals	Affiliate product tags
Which programme	Branded content policy	Shopping Affiliate
Status	Policy published, detection announced	A limited, running test
Paid Product Placement videos	Are the subject	Are excluded
Where it is documented	Two YouTube Help Center pages	Trade reporting only

What YouTube said it is running

Social Media Today quotes YouTube describing it directly: "We're running a small affiliate shopping experiment to show the exact same creator-tagged affiliate product, but sourced from a different eligible local retailer." The stated goal is to see whether auto-optimising which local merchant serves a product improves shopping performance. Same product, different seller, chosen for the viewer's location.

It excludes exactly the videos this post is otherwise about

The carve-out is explicit, and it is why the two announcements should never have been merged. Social Media Today reports that "the test will only relate to affiliate product tags via its Shopping Affiliate program", and MediaPost, reporting separately a day later, states that the test "will only apply to product tags within the shopping affiliate program" and does not touch videos where Paid Product Placement is enabled.

So branded content is the exception to the sourcing test, not its subject. A video with the paid promotion box ticked keeps the retailer the creator chose. The substitution only happens on affiliate tags in videos that are not declared brand deals.

Why this matters beyond a naming quibble

Because the two announcements point in opposite directions for the same person. The disclosure change adds an obligation to declared brand deals. The affiliate test adds automation to undeclared affiliate links. A creator who reads a merged summary can easily conclude that declaring a sponsorship will let YouTube swap their brand's retailer, which is the reverse of what YouTube published. Precision here is not pedantry, it is the difference between declaring and not declaring.

What changes if you make branded videos

Most of the practical consequence is not in the enforcement. It is in what the label does to a video that was going to be labelled anyway.

Disclosure stops being a post-hoc checkbox

The old workflow put disclosure at the end: film, edit, upload, remember the box. The announced detection inverts the incentive. Once a classifier can apply the label for you, the cost of forgetting is no longer zero and no longer deferred, and the cheapest place to handle it becomes the brief rather than the upload screen.

That is a production change more than a legal one. Four answers now belong on the same page as the script, because all four are fields somebody fills in at upload:

• Whether anything of value changed hands, including gifted product
• Which countries the video is cleared to run in
• The minimum age, overall and per country
• For a restricted category, whether the brand partner is Google-certified

Anyone who has run a multi-market creator campaign knows those answers currently live in three different threads and one voice note. The upload screen is the worst possible place to discover that nobody asked.

Two consequences you inherit with the label

The label is not only a badge. YouTube attaches two behaviours to it that change the economics of a sponsored upload, and both are on the policy page:

• Ad conflict avoidance. YouTube says "we may replace an ad that conflicts with your brand partner's ad with a different ad on videos with branded content", which it explains as protecting the sponsor relationship. Useful for the brand, and a variable in the creator's revenue on that video.
• Removal from YouTube Kids. The narrower one, and easy to miss: "videos with branded content will be removed from the YouTube Kids app in line with our existing policies". If your category is toys, snacks or family products, an automatically applied label is also an automatic removal from a surface you may have been counting on.

Both land on the creator's channel rather than on the brand, which is a reason to raise them in the brief. If you are already running ads from a creator's handle, the labelling state of the organic asset is worth confirming before any amplification budget is committed against it.

Viewers see it at the front, too. YouTube says "A disclosure will appear at the start of your video to let viewers know it contains branded content." The first frame of the video is now shared with the disclosure, which is an argument for hooks that survive a banner rather than hooks that compete with one.

Where AI-made UGC sits in this

The definition is about the relationship, not the camera. Branded content is content "influenced by a brand partner in exchange for something of value", so the question the policy asks about an AI-generated video is the same one it asks about a filmed one: did somebody give you something for making it. A video you generate for your own brand and upload to your own channel has no brand partner in it, and YouTube has not published guidance on that edge, so treat it as unsettled rather than as answered either way. A video an agency generates and a creator posts as their own is a brand deal, and the box gets ticked.

Worth noting what is not in scope here: YouTube says branded content is not held to the same technical bar as advertising, writing that "we do not require branded content to meet the same technical requirements" as ads. This policy governs uploads on a channel, which is a different surface from a media buy. If you are running paid placements, the rules on AI-generated images in Google Ads are a separate stack with separate consequences, and YouTube's own ad surfaces move on their own schedule, as an unannounced image ad format showed earlier this year.

Declare it before a classifier does

The useful reading of September 3 is not that YouTube got stricter. The obligation to disclose has been there for years and this update barely touched it. What changed is that non-disclosure stopped being invisible, and the machine that will make it visible has been announced without a date, which is the least comfortable state for anyone planning a fourth-quarter creator campaign.

So the move is unglamorous, and it fits on one page of a brief:

1. Put the sponsor question, the target countries and the minimum age next to the script, not on the upload screen.
2. Assume the wider of the two published scopes until somebody publishes the narrower one.
3. Treat the certification override as a fallback you would rather never test.
4. Check the restricted categories before you sign, because certifying the brand partner is your obligation, not theirs.

None of that is new advice for anyone who has been tracking how each platform labels AI-made ads, or who watched Google start showing viewers an AI label, or who has worked through TikTok's disclosure checklist. The pattern repeats: the obligation is old, the detection is new, and the gap between them is where the damage lives. And if your team is producing UGC-style ads with AI actors in Novoads, the disclosure decision still belongs to whoever posts the video and to the relationship behind it, not to the tool that rendered it. Novoads is $49 a month for the Starter plan with 50 credits a month, and the compliance work it removes is exactly zero, which is the honest thing to say about every production tool in this category right now.