A regulator that moved in stages
On 24 September 2026, Bangladesh's telecom regulator, the BTRC, approved Banglalink, the local unit of VEON, to launch a limited Starlink Direct-to-Cell (D2C) service, according to Tesla North's report of the approval. The service starts with SMS and data-based messaging on ordinary 4G handsets. It needs no dish and no special equipment. Banglalink calls it South Asia's first such service. Internet data over the satellite link is planned for a later phase, subject to further BTRC approvals.
The path to approval was incremental. A proof of concept began in May 2026 and ran for two months in Cox's Bazar. The BTRC proposed the trial, and officials and government agencies attended, according to The Financial Express. Banglalink then asked for commercial approval. The Financial Express noted that the regulatory decision would determine how D2C uses spectrum, where it can operate and how it interacts with terrestrial networks. Reports of an October commercial launch with a free first month come from secondary sources. The Week dates Starlink's announcement to 8 October 2026 and says the service is bundled with Banglalink rather than sold separately.
The case for caution
The strongest argument for a slower approach is sovereignty. A satellite signal that reaches handsets directly can bypass national gateways. That raises questions about lawful interception, spectrum interference with licensed terrestrial operators, and who is accountable when something goes wrong. Bangladesh's rules for dish-based satellite broadband reflect those worries. Under the NGSO guidelines published on 26 March 2025, operators must route user traffic through local gateways and equip their systems for lawful interception. The Business Standard reported that licences run ten years, with a mandatory five-year service commitment. Leaving early without sufficient justification can forfeit a $25 million deposit.
These are not frivolous conditions. A regulator that licenses a foreign constellation is entitled to know how it will be policed.
Why the staged approach is still the right one
The approval deserves credit for what it did not do. It did not wait for a comprehensive D2C statute, and it did not demand that the service be fully built before testing. Regulators observed a real trial, then approved a narrow first phase. Data and voice-like services wait for later sign-off. That is proportionate regulation: the risk-bearing features are held back until the regulator has evidence.
The benefit is concrete. The Week reports that Starlink targets seas, remote islands, coasts and hill tracts, and also outages caused by cyclones, floods and power cuts. In a delta country where storms regularly knock out towers, a text channel that works without a tower is a public-safety tool as well as a commercial product.
Bangladesh has also shown it can license satellite services quickly. The Daily Star reported that the BTRC granted Starlink its licence on 29 April 2025, after the NGSO guidelines of 25 March, with tariff approval still required before launch. The Week cites BTRC data putting Starlink's dish-based broadband at 3,469 subscribers as of January 2026. That is a modest base, and it suggests the price of dishes, not regulation, is now the main brake on satellite uptake. D2C removes the dish cost. That is why it matters more for inclusion than the broadband licence did.
The unresolved risk is surveillance, not spectrum
The real policy gap sits one layer down. The interception expectations attached to satellite licences operate in a country where, in the words of a 28 July 2026 joint statement by Access Now, Amnesty International, ARTICLE 19, Human Rights Watch and others, surveillance powers were expanded "often without meaningful transparency, independent oversight, or effective accountability mechanisms." The signatories call for surveillance powers to be created only through primary legislation, for prior authorisation by an independent court, and for parliamentary oversight. They also ask that the Personal Data Protection Act, 2026 narrow its broad exemptions for law enforcement, national security and intelligence.
This matters for D2C in a specific way. A handset that connects to a satellite when towers fail is most valuable in a crisis, and crises are when states are most tempted to demand access. If lawful interception obligations exist only in licence conditions and guidelines, users cannot know what can be demanded, by whom, or under what judicial check. Licence conditions can be rewritten administratively. Statutes cannot.
There is also a regulatory-design risk in the opposite direction. If every new satellite-to-phone feature needs a fresh round of discretionary approvals, the incentive is to bundle compliance with political goodwill. A published decision framework would help. It should say what the BTRC tests at each phase, how spectrum coexistence is assessed, and how long a decision should take.
What good looks like
Three steps would keep the benefits and reduce the risks.
- Publish the approval. The BTRC should release the conditions attached to Banglalink's phase-one approval, including spectrum terms and any interception obligations, so operators and users can see the rules.
- Put interception on a statutory footing. Any access to D2C traffic or metadata should require prior court authorisation under primary legislation, as civil-society groups have urged.
- Set phase gates in advance. Data and voice phases should have stated criteria, such as interference results and quality-of-service data from the Cox's Bazar trial, rather than open-ended discretion.
A comparison is useful. The US Federal Communications Commission authorised supplemental coverage from space in a partial grant to SpaceX released on 26 November 2024. It approved the T-Mobile arrangement and deferred requests for more power and more satellites. Staged authorisation is therefore not a Bangladeshi quirk. It is how serious regulators are handling this technology. Bangladesh's distinguishing feature is not the staging but the thin public record around it.
The approval is good news for people on islands and in the hills who have lacked coverage. The test now is whether the BTRC pairs speed with transparency, and whether lawmakers give users the protections that licence terms cannot.