A wholesale network that no longer has a public backstop
On 1 October 2026, Digital Nasional Berhad (DNB) completed a RM5.2 billion syndicated Islamic term financing. The deal refinances a government-guaranteed revolving credit facility that expired on 29 September. The same report says MCMC's spectrum assignment to DNB takes effect from 1 October and covers all 240 MHz of DNB's spectrum holdings. On 23 October, Light Reading reported that the Ministry of Finance was divesting its 31.18% stake. CelcomDigi, Maxis and YTL now hold equal shares, and the government keeps only a special share "to protect national interests."
This is a market-structure story, but it has a net neutrality dimension. Malaysia now has two 5G wholesale networks: DNB's, and U Mobile's own, after U Mobile's exit from its DNB wholesale arrangement took effect on 1 July 2026. With only two networks, internet access treatment is shaped less by retail competition than by what the wholesale contracts and the regulator's access rules allow.
A caveat on scope: we found no Malaysian net neutrality rule, ruling or consultation from the past 60 days. MCMC has not, to our reading, published a rule on traffic management or slicing. This piece is analysis of how existing access rules could carry that weight. It is not a report of a new neutrality development.
The strongest case for hard neutrality rules
The best argument for explicit rules is structural. A duopoly of wholesale networks has little competitive pressure to keep traffic treatment even. Slicing makes this sharper. A slice is a guaranteed-performance lane, and whoever controls slice allocation can in principle favour affiliated services. Europe took this seriously. The BEREC Open Internet guidelines require "equal and non-discriminatory treatment of traffic in the provision of internet access services" under Articles 3 and 4 of the Open Internet Regulation. BEREC has since put out a draft further guidance on 5G network slicing dated 5 June 2026. Critics of a light-touch approach have a point: voluntary restraint is weak when the incentives run the other way.
Why Malaysia's access rules are already doing part of that job
The answer does not have to be a European-style rule. MCMC's Review of Access List public inquiry report, dated 31 July 2026, keeps two 5G services on the Access List: 5G Standalone Access and 4G EPC with 5G RAN Access. It also records MCMC's view that excluding wholesale network providers from seeking access would be "harmful for retail competition." That matters because it keeps the radio network reachable by rivals, so an operator cannot easily lock a customer base behind one network's traffic policy.
The report's service description for 5G Standalone Access also lists "Network Slice Selection Function (NSSF and 5G Network Slices as selected by the Access Seeker)" among the facilities an access seeker may request. In other words, under the proposed text the party choosing the slices is the access seeker and not the network owner. The report also requires that access be supplied in line with 3GPP Release 15 and later updates. It adds that access must support non-public-network and machine-type services "to the extent technically supported by the Access Provider's radio network." That qualifier is a real limit, and regulators should watch how it is applied.
The structure is sound for a pro-innovation reason. It lets operators build differentiated products, such as enterprise slices and industrial IoT, without a licence-by-licence approval. It makes the terms of access, rather than a ban on prioritisation, the object of regulation.
Where the risk really sits: consumer-facing prioritisation
The sharper question is retail. U Mobile's own announcement describes "network prioritisation for eligible users during periods of congestion" and "reserved network capabilities." Those are lawful commercial offers, and paid tiers are normal in mobile. But a congestion-time priority tier is exactly where neutrality questions arise, because the line between selling users a better plan and degrading everyone else's depends on how the priority is defined.
Three principles would let MCMC protect users without freezing the product space:
- Differentiate by user or plan, not by content owner. Priority sold to a subscriber is a pricing choice. Priority sold to a particular application, or to affiliated content, is discrimination.
- Treat slices as specialised services, not as internet access. Enterprise and industrial slices should not reduce the quality of the general internet-access service. That is the logic of the European specialised-services carve-out, and it costs little to adopt.
- Publish the terms. Reference access offers and traffic-management policies should say how priority is allocated, so MCMC and rivals can check them.
What MCMC should do next
Malaysia has a regulator that has already chosen the less intrusive path of access regulation over content-level rules, and the reasoning in its own report points the right way. Because the report records MCMC's preference for open radio access and for slices selected by the access seeker, the cheapest fix is to write one clarification into the final instruments: access to slices and priority features must be offered on non-discriminatory terms to all access seekers and must not degrade general internet access. A mandatory standard on traffic management transparency could follow if complaints appear. Nothing here needs a ban.
The opposite risk is real too. If MCMC regulates slicing too tightly, the first to lose are enterprise and industrial users, the ones the 5G build-out was meant to serve. A DNB that has just raised RM5.2 billion to expand needs an access framework that is predictable and fair. It does not need a rulebook borrowed whole from another market.
The financing closes one chapter of state involvement in 5G. The open question is whether the rules that replace it are explicit enough to protect users and light enough to let operators build.