On 3 March 2026 Unity sent a notice to publishers in China, Hong Kong and Macau.
Their assets would stop selling on the Global Asset Store on 31 March. That is 28 days of warning.
Nobody broke a rule. Nobody got a strike. Unity's stated reason was updated regional licensing, distribution and compliance requirements for publishers in the Greater China Region.
What the notice actually said
Buyers keep what they already bought. Those assets stay in the My Assets list and stay downloadable.
They get no more updates and no support. Anything bought in the previous six months could be refunded, and a refund removes access.
Unity published a list of the assets coming down. 80 Level counted more than 3,000 of them. Buyers had to open that list and check their own library by hand.
Read the refund line twice. Taking the money back deletes the asset from your account.
If that asset is inside a shipped game, the refund is not a refund. It is a choice between your money and your project.
Why I cannot be calm about this
I am Iranian. I run a four-person studio in Budapest. I sell on ArtStation, BlenderKit, Superhive and my own shop.
Most people reading this are in Iran, Brazil, Turkey, Indonesia, India or Poland.
To a large platform we are not really a market. We are a compliance question somebody reviews once a year.
The engine went first
This was not the opening move. Unity 6 was pulled from those same three regions in April 2025.
Developers there were moved to a local engine called Tuanjie, built on Unity 2022 LTS.
So the engine left in April. The store followed eleven months later. That was the warning shot.
Open your payout page today
Here is the whole job, and it takes about ten minutes.
Open the payout page on every store you sell through. Write down what each one paid you last year. Divide your biggest number by the total.
That fraction is the only risk number that matters here. If one platform is more than half your income, a memo you will never see can cut your year in half.
I did mine this month. I am not going to publish the figure and neither should you. Not knowing it is the actual danger.
A good rate on somebody else's shelf
Unity pays publishers 70 per cent of revenue. That is a good rate. It is one of the better ones in this business.
The Greater China publishers were not on a bad deal. They were on a good deal, on a shelf that was never theirs.
That is the part people get wrong. Platform risk is not about how well a platform treats you today.
It is about who decides, and how long they have to give you.
Europe does the same thing, more quietly
ArtStation, February 2025. The EU Digital Services Act made every seller declare Trader or Non-Trader status.
Sellers who did not complete the form stopped being able to sell to buyers in the EU. Product pages stayed visible. The buy option simply was not there.
No ban, no email you would remember. A form went unfilled, and a continent stopped being able to pay you.
Four things worth doing this week
- Check which country each platform thinks you are in. A compliance review reads your tax form and your payout profile. It does not read your passport or your street.
- Download your own products from every store you sell on. Buyers in Greater China had to grab their files before the deadline or lose them. A library on a platform is a licence, not a folder.
- Keep a way to reach buyers that no store owns. An email list is yours. A follower count is not.
- Make one shelf your own. Mine is 358 products on my own shop, including the eighteen Blender add-ons I also sell on Superhive.
The shelf nobody can take
My own shop is slower than the marketplaces. It converts worse. It costs me hosting, code and support hours that ArtStation would cost me nothing.
It is also the only place where no compliance review can decide my region is a problem.
I have written before about a marketplace changing owner, and about a policy that moves work onto sellers.
Those are slow risks. You see them coming and you get to argue about them. This one is different.
It arrives finished, in an email, with a date on it.
Do it while nothing is happening
Twenty-eight days is not enough time to build an audience. It is barely enough time to export your listings.
None of this needs a plan. It needs one number, written down, before you need it.