The fintech sandbox inside Vietnam’s international financial centre contains the boldest provision in Resolution 222/2025/QH15: permission to pilot business models for which Vietnamese law has no provision, together with exemption from liability for implementing them in good faith. For founders and investors who have spent years asking Vietnamese regulators whether a product is permitted, that changes the question entirely.
What the fintech sandbox actually provides
Two things, and the second matters more than the first. The first is a controlled testing environment – Decree 323/2025/ND-CP directs the Da Nang location to serve as a controlled testing ground for new financial models. The second is the liability position: a pilot conducted within the sanctioned framework does not expose the participants to the consequences that would ordinarily follow from operating without a legal basis.
In most jurisdictions a sandbox merely relaxes licensing conditions for a defined cohort. Vietnam’s formulation goes further by addressing the regulatory-uncertainty risk that has been the real deterrent – not the cost of a licence, but the fear of retrospective enforcement against activity that no rule contemplated.
Which products the fintech sandbox is designed for
The priority list points clearly. Digital asset products and the platforms that trade them, covered in our digital asset exchange guide. Digital payment rails and settlement mechanisms. Specialised and novel exchanges. Supply chain finance platforms. Non-deposit-taking lending models. Third-party service providers supporting all of the above. The common thread is activity that sits outside the traditional banking and securities perimeters and therefore has no home in existing licensing categories.
How to approach a fintech sandbox application
Sandbox regimes reward specificity. Applicants should define the product narrowly, state the consumer and systemic risks candidly, propose the limits themselves – transaction caps, participant numbers, duration – and set out the exit path to full licensing if the pilot succeeds. Regulators grant space to teams that demonstrate they understand what could go wrong; they refuse teams whose application reads as a request for permission to do anything.
The governance package matters as much as the product. A named accountable executive, a documented risk framework, incident reporting commitments and a clear plan for winding down positions if the pilot ends are the elements that convert a promising idea into an approvable one. Our IFC membership guide sets out how this fits the wider licensing route.
Why Da Nang carries the fintech sandbox mandate
The Da Nang financial centre was given the experimentation brief deliberately: an innovation-linked ecosystem, digital technology, sustainable finance and a lower cost base for the technical teams these products require. Because the rulebook is uniform across both locations, a fintech sandbox licence obtained there carries identical authority to one issued in Ho Chi Minh City – a point our international financial centre overview explains.
Fintech sandbox FAQs
Does the liability exemption cover everything?
It covers good-faith implementation within the sanctioned pilot. It is not a shield for fraud, for operating outside the approved parameters, or for ignoring conditions imposed on the pilot. Teams that treat it as blanket immunity misread both the letter and the intent.
What happens when the pilot ends?
Successful pilots are expected to graduate into a licensed category, which is why the exit path belongs in the original application. From 2026 the centre’s institutions are mandated to propose new mechanisms – and pilots that worked are the evidence base for those proposals.
Can foreign founders participate?
Yes, through a member entity, with the employment and work permit facilitation covered in our employment guide. Framework texts are published via the Ministry of Finance.
What supervisors look for in practice
Reviewing a fintech sandbox application is an exercise in risk triage, and supervisors read for four signals. Whether the applicant has identified the specific harm its product could cause to consumers, counterparties or the wider system – not generic risk language, but the concrete failure mode. Whether the proposed limits are proportionate to that harm and demonstrably enforceable in the technology rather than merely promised in the application. Whether there is a named person accountable when something goes wrong, with the seniority and the authority to stop the pilot. And whether the applicant has thought about the end state, because a pilot with no graduation path is a permanent exception, which no regulator wants to create.
Teams that treat the application as a compliance formality tend to receive narrow permissions and heavy conditions. Teams that treat it as a genuine risk conversation – volunteering the uncomfortable scenarios and proposing their own mitigations – consistently receive broader scope. The asymmetry is well understood by firms that have run pilots in Singapore, the United Kingdom or the Gulf, and it applies with equal force here.
The final practical point is documentation discipline. Everything committed in the application becomes a condition in substance, so a pilot that quietly drifts beyond its stated parameters loses the protection that made the fintech sandbox attractive in the first place.
This article is provided for general information only and does not constitute legal advice for any specific case. Regulations may change – please consult a qualified professional before acting. Should further analysis be required, please contact one of our lawyers below.
