Norway's Autonomous Vessel Playbook: What AUKUS Nations Should Copy

Week 19 | April 2026

Norway has been running the world's most advanced autonomous vessel program in commercial service since April 2022. Not trials. Not demonstrations. Commercial operations, with cargo aboard, on a fixed route, certified by the Norwegian Maritime Authority (NMA).

The NMA did this without passing a single new law.

Meanwhile, the US Coast Guard has told Congress it lacks authority to waive statutory crew minimums for autonomous commercial vessels. Australia's regulator cannot approve alternative crewing solutions through its standard equivalence pathway. The UK is operating on case-by-case exemptions with no statutory framework in place for vessels over 24 metres. Three AUKUS nations, hundreds of billions in combined defence autonomous vessel investment, and the commercial side of each is effectively grounded.

What Norway actually did is worth understanding in detail. Not because it is an easy model to copy. But because the gap between what Norway built and what AUKUS nations have built is not a technology gap, or a funding gap. It is a governance gap. And governance gaps are entirely self-inflicted.

The vessel is a 3,200 tonne electric containership running an 11-nautical-mile shuttle between Yara's Heroya fertiliser plant and the deep-sea container port at Brevik. One hour each way, approximately 2.5 round trips per week. By early 2026 it had completed over 250 voyages and transported more than 35,000 containers, replacing roughly the same number of diesel truck journeys and cutting around 1,000 tonnes of CO2 annually.

Auto-crossing and auto-docking are operational, with onboard crew reduced from five to three. A further reduction to two is in progress, with the ship's electrician moving to Massterly's Remote Operations Centre (ROC) in Horten as part of a Chief-to-Shore project that began in November 2024. No publicly reported major incidents in the sources reviewed across four years of operations.

In March 2026, Massterly's ROC reached a milestone worth pausing on. A single technician began supervising three vessels simultaneously: Yara Birkeland plus two of ASKO's autonomous electric cargo barges on the Oslo Fjord. That is not three autonomous vessels operating independently. It is three vessels under coordinated remote supervision from a single shore-based centre, which is a different and more precise claim. But it represents the operational model Norway has been building toward since 2018: one qualified person ashore, multiple vessels at sea, technology handling the navigation.

The Massterly Managing Director framed the purpose clearly: autonomy is the means, not the target. The target is a sustainable, competitive maritime industry that can survive in a high-cost labour market. Norway experienced 1,500 cancelled ferry crossings in a recent year due to crew shortages. Autonomous systems are not a futurist project, they are a staffing solution.

Norway's framework rests on a single piece of primary legislation: the Ship Safety and Security Act (Skipssikkerhetsloven), Act No. 9 of 16 February 2007. Three provisions within it do the work.

Section 9 requires ships to be designed and equipped to provide satisfactory protection of life, health, property and the environment. Section 14 requires safe navigation and Section 15 requires the ship to be safely manned with adequate watchkeeping arrangements. None of these provisions specify how many crew are required, or that a human must be physically aboard. They specify outcomes. The mechanism that makes autonomous operations permissible is the equivalence principle: if an automated system can demonstrate it performs a required safety function to at least the same standard as a human crew member, the requirement is met.

The NMA formalised this into an operational framework through Circular RSV 12/2020, issued in August 2020. This circular governs the construction and operation of all autonomous and remotely operated vessels on Norwegian domestic voyages. It requires a structured documentation package from operators: Concept of Operations, Safety Philosophy, Design Philosophy, GAP Analysis against existing regulations, full Risk Assessments and HAZID, FMEA, third-party verification (typically DNV), a manning analysis, and a Safety Management System. The process is rigorous. It is also transparent, consistent, and predictable.

Central to RSV 12/2020 is the Minimum Risk Condition concept: at least two fallback states must be available during normal operation, at least one of which survives fire, flooding, blackout, or loss of communications. This replaced the binary question of crew or no crew with a functional safety framework. Operators are not asked whether a human is present. They are asked whether every required safety function has a robust, documented, verified alternative.

The physical infrastructure came first. In September 2016, the NMA and the Norwegian Coastal Administration jointly designated the world's first official autonomous vessel test area in the Trondheim Fjord. A second test area followed near Horten in December 2017.

On April 9, 2026, a decade after that test area opened, Maritime Robotics ran a live autonomous drone swarm demonstration for the US Navy in that same Trondheim Fjord, showcasing coordinated multi-vessel operations, remote supervision, and real-time control in a certified environment. That is what regulatory infrastructure investment produces: an ecosystem capable of demonstrating allied capability in a certified test area, on demand. Testing in a designated area is a prerequisite for operational approval under RSV 12/2020. Norway built the sandbox before anyone asked for it. Testing in a designated area is a prerequisite for operational approval under RSV 12/2020. Norway built the sandbox before anyone asked for it.

None of this required a Storting vote. No parliamentary legislation was passed for autonomous vessels. The 2007 Act's goal-based structure already gave the NMA the room to approve innovation administratively. What the NMA then did was staff its dedicated New Maritime Technology team with approximately seven people and adopt an institutional posture of co-development with industry rather than gatekeeping. NMA staff participate directly in HAZIDs. They review documentation at each phase. When a ship owner tested successfully and demonstrated no safety incidents, the NMA allowed them to reduce crew from five to three. Then toward two. The process is iterative, evidence-based, and governed by a consistent national framework rather than ad hoc case-by-case assessments at the local level.

In early 2024, Norwegian ferry operator Fjord1 signed a contract with Tersan Shipyard in Turkey for four autonomous zero-emission double-ended ferries, targeting the Lavik-Oppedal route across the Sognefjord. Steel cutting began in October 2024. All four vessels are scheduled for delivery in the first half of 2026, with conventional crewed service beginning in September 2026, auto-crossing and auto-docking functions implemented from January 2027, and full autonomous navigation monitored from a shore-based control centre in Floro targeted for 2028.

The programme involves HAV Design for naval architecture, Norwegian Electric Systems for the autonomous navigation scope, and Schottel for propulsion. The vessels are 120 metres, carry 399 passengers and 120 cars, and are fully battery-electric. The contract period runs to 2034. The route carries 1.2 million cars annually on the E39 highway corridor.

This programme is more likely to hit its timeline than Yara Birkeland hit its original targets, for a specific reason: it benefits from an established regulatory pathway that did not exist when Yara Birkeland began. RSV 12/2020 was published in 2020. The classification framework is mature. Kongsberg Maritime has already completed over 20,000 automated crossings on another Norwegian ferry route. The phased approach builds in buffer time between crewed operations, partial automation, and full autonomous navigation. The 2028 target should still be treated with informed caution, particularly given the added complexity of passenger-carrying certification, but the regulatory infrastructure that blocked or delayed earlier programs is no longer a constraint.

The question that kills autonomous vessel programmes in most jurisdictions is labour. Norway answered it by framing the problem differently.

From January 2026, the NMA began certifying a new category of shore-based maritime professional: the Vessel Manager, licensed with Radio Operator Certificate and General Operator Certificate equivalents, responsible for coordinating navigational authority and system integrity monitoring from a Remote Operations Centre. This is not a replacement job description dressed up with a different title. It is a certified maritime career pathway that did not exist five years ago, that was created in parallel with the technology, rather than in response to it.

Crew shortages helped. When an industry faces 1,500 cancelled departures a year because it cannot staff vessels, unions and operators share a problem. Autonomy positioned as a solution to a shared problem lands differently than autonomy positioned as a cost-cutting exercise. Norwegian ferry operators made their crew integral to technology development from the start. One operator was direct about it: consultation with captains and crew has been crucial. This is an aid, not a replacement.

The Norway Ocean Autonomy Cluster, now at 70-plus members and renewed with three-year strategic support in October 2025, is managed by the same body that runs Norway's Defence and Security Cluster. That institutional overlap is not accidental. The maritime autonomy ecosystem Norway is building has explicit dual-use architecture baked in. More on that in a future edition.

The contrast is structural, not incidental.

In the United States, the core barrier is statutory. Multiple provisions of Title 46 of the US Code establish minimum crew requirements that the Coast Guard cannot waive. Section 8101 prescribes the complement of inspected vessels. Section 8102 covers watchmen requirements. Section 8104 covers watches. Chapter 83 governs licensing of masters and officers. These provisions assume a human is aboard and in control. The GAO documented this bluntly in testimony to the House Subcommittee on Coast Guard and Maritime Transportation in December 2025: the Coast Guard currently does not have authority to waive these requirements for autonomous vessels outside the limited scope of a single pilot programme for at-sea rocket recovery. The agency also told Congress it is not intending to issue domestic guidance ahead of the IMO framework, meaning US commercial maritime regulation will lag even the non-mandatory IMO MASS Code, now due for adoption at MSC 111 in May 2026.

The demand is there. Since 2024, local Captains of the Port have received 48 requests involving autonomous ship technology, each handled case-by-case at the local level. The Coast Guard itself acknowledges that industry has raised concerns about consistency across ports. That is what happens when you have 48 individual negotiations instead of one national framework.

Australia's position is similar in effect if different in mechanism. The Navigation Act 2012 and the Marine Safety (Domestic Commercial Vessel) National Law Act 2012 subject autonomous vessels to the same framework as conventional crewed vessels. The critical barrier is that the Australian Maritime Safety Authority cannot grant equivalent means of compliance relating to crewing through its standard pathway. Operators cannot propose alternative crewing solutions and have them assessed against a functional safety standard. They need specific exemptions, navigating what an independent review described as a process that is inefficient, opaque, and uncertain. The review recommended amending both Acts. As of April 2026, no legislation has been passed.

The UK is the most advanced of the three. The Workboat Code (2023) includes provisions for remotely operated vessels under 24 metres. The Maritime and Coastguard Agency issued a general exemption in August 2024 for autonomous vessels under 2.5 metres. The industry Code of Practice for MASS is now at its sixth edition. A September 2023 regulatory review proposed amending the Merchant Shipping Act 1995 to address autonomous vessels. That review's proposals have not yet been enacted as a MASS-specific framework. The MCA handles larger autonomous vessels through case-by-case exemptions, the same approach it acknowledged creates an increasingly complex burden as vessel size and autonomy level increases.

The core failure across all three nations is the same: their maritime statutes use prescriptive language that assumes crew are present, and their regulators lack authority to approve functional equivalents without legislative change. Norway's 2007 Act used goal-based language - seemingly deliberately. The NMA then built the administrative apparatus to use that flexibility. AUKUS nations wrote prescriptive statutes, and none have thus far changed them.

Here is the uncomfortable addition to all of this. Even if AUKUS nations accelerated today and adopted Norway's regulatory model in full, Norway is no longer the frontier.

In early April 2026, Japan’s ClassNK granted its autonomous navigation notation to the container ship Genbu, a 696 TEU vessel built from the keel up for autonomous operations, now in commercial service on the Kobe-Yokohama corridor as part of Japan's MEGURI2040 programme. This is the world's first autonomous navigation notation for a vessel on medium- to long-distance coastal routes, not the short controlled shuttle Yara Birkeland operates.

South Korea's HD Hyundai Avikus received full Type Approval from DNV on April 7, 2026 for its HiNAS Control autonomous navigation support system, the first mass-produced autonomous navigation system to receive international certification applicable across multiple vessel types. Type Approval means HiNAS Control can now be installed on vessels without additional per-project verification. Avikus already has over 500 orders for the system, and it has been standard specification on all HD Hyundai new-builds since 2023.

South Korea also has something none of the other leading nations have: a dedicated statute. The Autonomous Ship Act, Act No. 19909, entered into force on January 3, 2025, providing for five-year master plans, designated autonomous ship operating areas, and safety assessment requirements. It is the first national legislation in the world written specifically for autonomous ships.

Norway pioneered the regulatory sandbox and used it to build a functioning ecosystem. Japan and South Korea are now deploying at commercial scale with dedicated classification frameworks and, in Korea's case, dedicated legislation. AUKUS nations face two reference points simultaneously: Norway's model to copy, and Asia's pace to match. Both benchmarks are moving.

The US paradox is the sharpest version of a problem all three AUKUS nations share. The Navy is building autonomous warships, funding Ghost Shark production, running AUKUS Pillar 2 autonomous systems programmes. The commercial sector waits. Not because the technology is unready. Because the statutes were written assuming a human would always be aboard, and nobody has changed them.

Norway did not solve this with money or technology. It solved it with a 2007 Act written in functional rather than prescriptive language, a circular issued by a seven-person regulatory team in 2020, two test areas designated before anyone asked for them, and an institutional culture that treats industry as a development partner. None of that is technically difficult. All of it requires political will and institutional design.

South Korea answered the same question differently: pass a dedicated statute. Japan answered it differently again: change six laws at once through a single omnibus act, funded by a private foundation, coordinated through a 51-company consortium. Three countries, three mechanisms, one outcome. Operational autonomous vessels in commercial service, certified, scaling.

AUKUS nations have world-class autonomous vessel technology in their defence procurement pipelines. The question for procurement officers and maritime lawyers reading this is a narrow one: is there a political appetite to do for commercial maritime regulation what Norway did in 2007? Because the technology is not the constraint. It has not been the constraint for some time.

Next week we go inside one of the more consequential technology gaps in AUKUS defence procurement: what happens when autonomous vessels, submarines, and unmanned systems need to navigate in environments where GPS is jammed, spoofed, or simply unavailable. Quantum navigation is not a future capability. It is being tested now, with Australian government funding, in Washington DC. We look at what QuantX Labs actually demonstrated, what AUKUS Pillar 2 quantum programs are genuinely delivering versus what is still hype, and why getting this right is the foundational requirement for everything else in maritime autonomy to work at scale.

Since you have been, thanks for reading.

Cheers,

Mick

Ocean Tech Intelligence provides informational analysis only. Nothing in this publication constitutes financial, investment, legal, or strategic advice. Readers act on this content at their own risk. For full details see our Disclaimer.