Second Autonomous Shipping Conference – Public International Law Aspects

The UCL Autonomous Shipping Project is pleased to announce that it will be hosting, in collaboration with the Institute of International Shipping and Trade Law of Swansea University), a Conference exploring the ‘Public International Law Aspects’ of the autonomous shipping revolution. It will take place on 6 April 2022.

This Conference follows from the success of the first Conference on ‘Liability and Contractual Issues’. The paper which Julian Clark presented at this event can be found here.

This second Conference will explore the complex public (international) law and technical issues surrounding the use of autonomous vessels at sea. This is significant because most international (shipping) conventions and domestic legislation have hitherto taken it for granted that all vessels have a master, officers, and crew on board.

The Speakers and Chairs at the ‘Public Law International Aspects’ Conference will be:

  • Dr Lia Amaxilati (IISTL, Swansea University)
  • Professor Simon Baughen (IISTL, Swansea University)
  • Dr Richard Caddell (Cardiff University)
  • Frederick Kenney (Director, Legal Affairs and External Relations Division, IMO)
  • Dr Youri van Logchem (IISTL, Swansea University)
  • Dr Melis Ozdel (UCL)
  • Cdr Caroline Tuckett RN (Naval Legal Services)
  • Tom Walters (Partner, HFW)

If you wish to book your place at the Conference or to find out more about it please click here.

‘Is a self-driving car smarter than a seven-month-old? Or, how to ask the wrong question to get to an irrelevant answer.’ by Dr Frank Stevens

In The Economist of 4 September 2021, an article was published under the title ‘ Is a self-driving car smarter than a seven-month-old?’.  If you guessed, from the way the question is phrased, that the answer was ‘no’, you would have been entirely correct. Ultimately, the article (intentionally or not) puts out an implicit message that these self-driving things are dangerously immature and should not be allowed on our roads.

The author’s point is that human children, generally around the age of seven months, have learned the concept of ‘object permanence’.  They realise that an object that disappears from sight is nevertheless still there.  A toy that is covered by a blanket isn’t gone, it simply isn’t visible anymore for the time being.  A self-driving car on the other hand doesn’t ‘know’ or ‘realise’ that a cyclist who disappears behind a parked van is actually still there and will re-appear momentarily from behind the van.  Hence, the author’s perceived need to improve the ‘intelligence’ of self-driving cars.

There are a number of issues with this approach and the (implicit) message that this article puts out.  For completeness’ sake, it’s not a given that self-driving cars do not apply the concept of object permanence or object tracking.  There are many different systems out there, the inner workings of which are not always entirely known to anyone other than the development team.  But even assuming that self-driving cars indeed do not know the concept of object permanence, the question and the comparison is still wrong.

The end-goal of self-driving cars – or other AI applications such as the software that beats Chess or Go champions, speech-recognition software, medical diagnosis software, etc. – is not to create a real-world C3PO or a robot that can carry on a witty conversation with Will Smith.  A self-driving car is a specialised system, built to perform a specific task: go from A to B, while complying with the traffic rules and without crashing into someone or something.  Obviously, self-driving cars today are not perfect (yet), but the question is not how they compare to young children.  Computers are able to process millions of instructions per seconds, but even computers do not have unlimited processing powers.  A three-year old will be able to tell you whether the animals in the field next to the road are cows or sheep, whether the headlights of an oncoming car are round or square, and whether the moon is out at night.  Your self-driving car will not be able to do any of these things – because there is no point in a self-driving car to be able to do so, and it would be simply wasting its processing power if it were to do so.  With object permanence, the question is not whether a seven-month-old has acquired this concept, the question is whether implementing this concept results in a better system.  Is the cost in processing power set off by the improvements in efficiency of the system?  It is by no means certain that this is the case.  The technical paper by which the article in The Economist was inspired[i] reports a 5% increase in tracking performance.  Tracking performance is only one of the factors in the overall performance of a self-driving car, and maybe, if the processing power required to implement object permanence were applied to other factors (forward scanning, speed of reaction, etc.), the overall performance of the self-driving car would improve more than by implementing object permanence.  After all, object permanence is not always relevant.  The cyclist who disappears behind the parked van may stop there, because that’s where he lives, or he may turn into a bike lane, and thus never re-appear from behind the van.  In such cases, tracking his possible trajectory is just a waste of resources.

The correct question, therefore, is not how we can make the self-driving car more “intelligent”, but how we can make this task-specific system more efficient – in every possible meaning of the word.  Actually, the road infrastructure today is limited, because it has to work with the limitations of human drivers.  Traffic lights, for example, are visual signals, which only work within line-of-sight.  Even the most proficient and attentive human driver doesn’t know what the traffic light around the corner is showing.  Traffic lights might be obscured by fog, overhanging branches, etc.  In a digital world, traffic lights can broadcast their information –data signals are not stopped by fog and can go around corners.  Your self-driving car of the future will know whether the traffic lights located two intersections down your path are red or green.  Self-driving cars can broadcast their ‘intentions’, i.e. the route they will be taking in the next few minutes, to the traffic around (ships today already do so (to a limited extent) with AIS), local repeaters can greatly increase the accuracy of GPS positioning, etc.  Again, with all of these factors, the question is not how to make self-driving cars more “like humans” or more “intelligent”, the question is how we can improve the efficiency of the system.  The same applies to any autonomous vehicle, be it on land, on water or in the air.

All of these are, of course, primarily design and technical issues, but they should matter to lawyers too.  Autonomous vehicles will not be perfect and accidents will happen.  Even the aviation industry, which by now has decades of experience with automation and computerised systems, still has accidents.  The Boeing 737 MAX scandal was a very sad reminder of that.  Self-driving cars and autonomous vessels will not have to relearn all of the lessons that aviation has learned over the years, but nevertheless, it is unrealistic to expect that everything will work as expected right from the start.

When accidents happen, lawyers immediately start thinking about negligence, fault and liability – as they should, of course.  But if their mind has been skewed before by news stories about making autonomous vehicles more like seven-month-olds, chances are that their appreciation of negligence will not be correct.  It goes without saying that designers and developers can indeed go wrong, but their performance should be measured against the correct yardstick, and not against a misleading and incorrect yardstick.

Dr Frank Stevens is an Associate Professor at Erasmus School of Law, Rotterdam.


[i] Jakob Suchan, Mehul Bhatt & Srikrishna Varadarajan, ‘Commonsense visual sensemaking for autonomous driving – On generalised neurosymbolic online abduction integrating vision and semantics’ (2021) 299 Artificial Intelligence 103522

‘On the Regulatory Challenges Linked to MASS’ by Professor Henrik Ringbom

The regulatory challenges relating to Marine Autonomous Surface Ships (“MASS”) involve two main areas of interest. First, there are the public law rules, covering safety, manning, technical and operational standards etc. for ships. These are the rules that determine whether it is lawful to operate MASS in the first place. Second, there are the liability rules which enter the scene when things have gone wrong. A key question here is how the increased automation in shipping may impact the distribution of liability among the parties involved, including entirely new players, such as software developers and providers of situational awareness (lookout), navigation and communication technology for MASS.

Uncertainty surrounding liability rules does not necessarily prevent the introduction of MASS. The backbone of the existing maritime liability regime, i.e. the owner/operator of the ship, does not disappear with increased automation, nor does the flag state. The novel liability issues that MASS introduce, such as the attribution of fault to computer systems or the scope liability rules governing errors committed by technology providers with only remote links to the ship, need to be settled over time, notably to ensure that the new developments do not represent unreasonable risks to third parties. However, it is neither necessary nor realistic to achieve clarity and consensus on these matters before such ships are in operation. This means that such matters will probably be accommodated gradually, once there is more experience with the practical operations of MASS, including through national standards and/or case-law-based interpretation. Gradual development, in turn, implies national and regional variations, which may necessitate discussion on whether a new international instrument is needed.

It could well be that the difference between MASS and other ships in this regard has been overemphasised. Are failures of MASS technology really that different from other technical failures on board ships (which are routinely handled by the existing legal system in all parts of the world)? Most analysts appear to answer that question affirmatively and call for some form of strict liability regime for the ship operator to fill the legal voids. A related question that has received less attention is whether the identified differences are limited to MASS. Should one not ask instead whether the increased reliance on technology on board ships justifies a closer look at the general liability rules for ship operations, and the range of persons included, rather than the creation of a (necessarily random) legal differentiation between MASS and other ships?

By contrast, there must be a good level of certainty and consensus on the applicable safety requirements before MASS can be brought into operation. This is particularly so for MASS engaged in international trade, but some of the key international rules apply in any sea area and it is not obvious that the territorial sovereignty of the state concerned will always trump such international commitments. The place to resolve questions related to safety requirements is the IMO, as international technical rules will make it significantly easier for other regulatory lawyers to follow suit. This applies ‘upwards’, to the jurisdictional regime governing flag and coastal states’ rights and obligations laid down in the 1982 UN Convention on the Law of the Sea, as well as ‘downwards’ in relation to national legal systems.

There is already a considerable amount of research and literature highlighting the legal difficulties involved in accommodating MASS into the existing legal framework. Maritime lawyers from all quarters, including academics, civil servants, and the IMO itself through its recently completed ‘regulatory scoping exercise’, highlight the differences of MASS as compared to traditional ships and the difficulties that arise from the wording of the existing conventions. A common assessment appears to be that a few amendments, some adjustments, and several general clarifications are needed, but that MASS generally give rise to few direct conflicts with the existing rules (the main exception being the watchkeeping provisions of the STCW Convention). Despite the decidedly human-oriented focus and background of today’s international safety rules, authorization of MASS would accordingly not require a major overhaul of the existing rules.

Oddly, little attention has been given to the question of how the numerous regulatory voids that MASS have unveiled should be filled, which arguably represents a much greater regulatory challenge than the adjustment of existing rules. Current rules on lookout, for example, are not easily translated into purely technology-based situational awareness information, which will be needed at any degree of autonomy or manning reduction. Machine-based decision-making in navigation is another regulatory blind-spot. For these and other new areas for regulation, standards will be needed to modify the current human performance-oriented rules into a new quantifiable format, but also to develop common rules and principles for data handling, communication, redundancy systems and emergency procedures, to name but a few.

The goals and content of such rules will have to be specified internationally in due course. However, the existing IMO framework already offers a range of mechanisms and procedures that could be used for advancing MASS developments. Most of them operate through flexibility offered to the ship’s flag state administration, and all are based on the principle that the overall safety level must not be compromised.

There is, for example, already an established systems for exceptions, alternative designs, and equivalent solutions for technical standards. This may prove an important avenue for flag state administrations to approve variations to technical standards, notably those laid down in the first three chapters of SOLAS and the Load Lines Convention.

Another agreed procedure currently in place relates to the approval of alternative designs, where accepted, through a complex dialogue between the owner and the administration, as outlined in IMO Documents MSC.1/Circ. 1455 and MSC.1/Circ.1212. This procedure might usefully form a basis for the approval of goal-oriented technical construction and equipment requirement for MASS too.

For operational requirements there are fewer exemptions or alternatives available. Generally the rules that regulate human activities on board ships, including the COLREGs, cover all ships and include limited scope for exceptions. However, the assessment of the safe manning of individual ships laid down in SOLAS Regulation V/14 and associated guidelines, includes a general appraisal of all functions required to be performed on the ship in question. This procedure could serve as the centrepiece for the assessment of the operational aspects of MASS, probably without much modification. Yet, the (national) criteria on which the assessment is based, in particular with respect to how technology solutions may assume duties traditionally performed by humans, need to be adjusted in the case of MASS.

At company level, the existing safety management procedure, as regulated in the SOLAS Chapter IX and the associated ISM Code, will no doubt also retain a crucial role in the management of safety for MASS. The current regulatory system appears well-suited for MASS, but the manner in which the procedures operate, e.g. on the communication between ship and shore, will have to undergo important alterations in the process.

The acceptance of MASS also requires considerations by relevant coastal states regarding whether a particular area is suitable for MASS operations, whether the safety of other ships and the environment is ensured, and regarding the availability of various supporting infrastructure etc. There is no precedent for handling such issues within IMO, but the 2019 Interim Guidelines for MASS Trials (MSC.1/Circ. 1604) represent a first step in this direction. The Guidelines may be interpreted as a cautious endorsement by the IMO of experiments with MASS in international areas, even if not all technical requirements are strictly met. In this sense, para. 2.2.1 of the Guidelines provides that “compliance with the intent of mandatory instruments should be ensured” (emphasis added). This endorsement, in turn, may have implications on whether MASS trials undertaken along the lines foreseen by the Guidelines (which are not limited in time or to a certain sea area) are perceived as being undertaken in accordance with “generally accepted rules and standards”, as required by the law of the sea. As far as watchkeeping is concerned, the Guidelines may have a more direct regulatory impact if they were considered to enable (constrained) exemptions from the watchkeeping rules under Regulation I/13 of the STCW Convention, which refers to trials conducted “in accordance with guidelines adopted by the (IMO)”.

Overall, these examples highlight that, even if there is a long way to go before regulation of MASS is anywhere near robust, the existing regulatory regime offers a variety of possibilities for individual states to support the development of MASS without having to await a full review of the IMO rules, and without being in conflict with the existing international rules.

Clarity on liability rules is desirable, but not a conditio sine qua non for operating MASS. This is different with respect to safety requirements, where the legality of operating MASS needs to be confirmed beforehand. In both areas, states may already prepare for MASS and have a number of mechanisms and instruments at their disposal to accommodate MASS into their legal framework, and to start specifying the key requirements that will be needed for such ships. In view of this, it is surprising that the regulatory debate on MASS to date has been so ‘backward-looking’ focusing on the impact on existing rules. The bigger challenge is arguably to agree on what rules and standards are missing, and how the gaps should be filled, in the shorter and longer term, nationally and internationally.

Professor Henrik Ringbom is a Professor II at the Scandinavian Institute of Maritime Law in Oslo, Norway, and Head of Research at the Department of Law, Åbo Akademi University in Turku/Åbo, Finland. He recently co-edited the book Autonomous Ships and the Law (Routledge, 2020).

‘A realistic approach to the development of autonomous ships and the law’ by Professor Johan Schelin

An expert on artificial intelligence once said that nothing is as difficult as trying to foresee the future, especially when it comes to technological developments and their significance. In the late 1960 and early 1970s people were very fascinated by space and landing on the moon. Some experts even predicted that in the near future ordinary cars would be replaced by small personal rockets. However, few experts foresaw that fifty years later people would devote much of their time to a small device called a smart phone, chatting and looking at media published on social platforms like Facebook, Instagram, Snapchat and Tik Tok. There seems to be a tendency to both overestimate and underestimate technological developments. Most likely we will see the same thing with the development of autonomous ships. Some experts predict a future where goods will be carried on board totally unmanned ships and goods will automatically be loaded and discharged in automated harbours.

Although we may very well develop the technology for this, other factors such as the capital and labour costs, ship and cargo safety, available infrastructure and acceptance by the public are important when it comes to the commercial use of autonomous ships. Unlike with the operation of autonomous ships for military purposes, a commercial ship owner will always have to take these factors into account.

Remotely controlled or fully autonomous ships will most likely be expensive to operate. Today a lot of maintenance work is carried out by people working on board while the ship is at sea e.g. parts of the machinery are replaced, the deck is painted. A remotely controlled or fully autonomous ship will instead have to be brought to a ship yard or a dock for such maintenance. Circulating a maintenance crew between autonomous ships in a company fleet would also be difficult since equipment on board one ship may break down suddenly at sea while the crew is working on board another ship. In addition, ships are often one of a kind structures which means that the maintenance work on board requires specific knowledge of the ship in question. It is, of course, possible to argue that in the future ships may be constructed in a way that minimises the maintenance required or that such work could be carried out by robots. However, materials that require no or little maintenance and advanced robots that can carry out many different tasks on board (if available at the market at all) tend to be expensive. In the end, all investments of this kind must be weighed against the benefits of human labour from countries with low labour costs doing the same work.  

In addition, a ship owner must consider that the sea is an environment of a very dynamic character. Hurricanes, storms and ice may affect the ship during the voyage. Equipment may suddenly break down, a storm may cause a displacement of the cargo, the main deck may get covered with heavy ice such that the ship is at risk of capsizing. These are risks that can be avoided with a crew on board, but which are difficult to manage on board a remotely controlled or fully autonomous ship. Also, the traffic situation must be considered. Parts of a ship’s journey often take place in sea areas with heavy traffic. If a fully autonomous ship causes a collision as a result of a deficiency in the navigational equipment there is a risk that not only the ship owning company will be liable for the damage but also the manufacturer of the equipment (due to product liability). It remains to be seen whether manufacturers are willing to bear that risk.

Remotely controlled and fully autonomous ships will also be dependent on a developed port infrastructure. This is illustrated by the so-called m/v Yara Birkeland project in Norway. Yara Birkeland is a fully autonomous ship which operates between two ports specifically designed for the purpose. However, many ports, especially in developing countries, lack such infrastructure. This makes it more difficult to operate remotely controlled or fully autonomous ships in the tramp trade.

Another factor of importance is whether unmanned remotely controlled or fully autonomous passenger ships will be accepted by the public. Even during short ferry trips, events may occur on board that require the crew to act e.g. passengers may fall over board or get injured. In comparison, it is possible today to technically construct a remotely controlled or fully autonomous aircraft, but at the same time there is a risk that many passengers would be hesitant to board such an aircraft. If something goes wrong, people still want a pilot to be there.

In light of the rapid development of autonomous ships, the International Maritime Organization (IMO) is now reviewing the international legislation on the manning of ships. Considering all the factors discussed above, it seems that a more realistic legislative approach at the international level at this stage would be to give priority to the development of new rules regarding periodic unmanned bridges rather than discussing rules on remotely controlled or fully autonomous ships. It also seems that tomorrow the need for humans on board most ships in international trade will remain.

Professor Johan Schelin is a Professor of Maritime and Transport Law at Stockholm University and is the Director of the Axel Ax:son Johnson Institute for Maritime and Transport Law.

‘Carriage of Cargoes – Ports and Autonomy’ by Julian Clark

A paper presented by Julian Clark at the ‘Autonomous Shipping – Liability and Contractual Issues Conference’ (27 October 2021, co-hosted by the UCL Autonomous Shipping Project and The Institute of International Shipping and Trade Law of Swansea).

I would like to start by asking all those present to raise their hands if they believe that within their lifetime they will see the regular use of automated commercial vessels and automated ports operating in the global supply chain regularly.  The next question that I wanted to ask you was whether any of you were familiar with the film “The Matrix”?  If you are not, then this next part of my presentation may fall on stoney ground.

In the film ‘The Matrix’, Orpheus provides Neo with this option:

“You take the blue pill and the story ends here, you wake up in your bed and believe whatever you want to believe.  You take the red pill, you stay in wonderland, and I show you how deep the rabbit hole goes”.

So I invite you to take the red pill and join me on a journey, for automation of ports is most certainly a very deep rabbit hole. 

I am going to take you on this journey in three stages.

(1) Where are we now and what is in store over the next 20 to 30 years?

(2) What are the non-legal challenges?

(3) What are the legal challenges?

(1) Where are we now?

In 1956 Malcolm McLean devised the concept of the shipping container.  Eight years later the world’s first fully purposed container vessel, MV Kooringa, set out on its maiden voyage. What followed was the greatest change in maritime transportation since the transition from sail to steam. 

If, and I am underlining the word “If”, the envisaged automation of vessels and ports comes to pass, I would submit that the change as a result will eclipse both the move to fuel based propulsion and containerisation.

For a vision of the future, I would invite you to Google when you get home “Kalmar Global’s 2060 Vision”. What they set out in a short video clip is a vision of a true on-demand logistics global supply chain, utilising the full automation of both vessels, road, rail freight and ports.  Where end-to-end product selection and mode of transport is geared in response to consumer demand. Where AI is utilised in order to optimise efficiency, and drone technology is an integral part of the maintenance regime.  Now this may at first glance appear as distant from reality as my opening quote from ‘The Matrix’, but I would suggest that this vision of the future is not that far away. 

During London International Shipping Week, Michael Parker, the chairman of Citigroup’s Global Shipping Logistics and Offshore business, provided a well-argued thesis that one of the reasons why the all-embracing concept of ESG and a move to carbon neutrality will succeed is due to the fact that consumer demand will both drive and fund such a change. As he said, “this is perhaps the first time in history where shipping can really take its place at the top table”. He argued (and I would support this) that in an on-demand society consumers will not only select greener forms of transportation but be willing to pay extra for them.

And some of what the Kalmar vision predicts is already here. The Box Bay system in Dubai is a fully automated container stacking system. London Gateway has 12 of the world’s largest ship to shore cranes. These cranes are 138 meters in height and can move 4 x TEU’s, or 2 x FEU’s in a single lift. Their remote operation means that they can function in virtually any weather condition. These leviathans are supported by 60 automated stacking cranes, able to load and unload 1800 trucks per day.

The ports of Newcastle and Singapore have already started to move to automation with significant investment.  Although what has to be the most impressive automated port in the world so far is that of the Maasvlakte II port in Rotterdam which can already process 2.7 million TEUs and by its full operation in 2030 is predicated to be able to meet 4.5 million TEUs. The estimate of global investment in port automation is currently running at around US$10 Billion and this figure is expected to double (or even treble) within the next five years. The Chinese port of Caofiedian is already reporting a 70% increase in efficiency and a 30% reduction in labour cost as a result of automation. And not only is large scale investment the driver here but the recognition of the need to move from a micro supply chain ecosystem to a global autonomous logistics network. Where we are likely to see the increased use of a combination of AI, machine learning, and Internet of Things technology, utilising automated equipment, robotic loading and discharging systems, drone maintenance bots, remote driverless vehicles, all controlled from a central terminal control tower or shore control centre. These centres would run both the logistics operations and the interfacing with vessels and consumers.

It is developments such as these and the real possibility of the introduction of virtual reality remote working and integration of smart containers (already brining a 20 to 40% improvement in cost and sustainability), which has led to those most closely involved in the operation and management of ports launching their own studies.

The UK Government’s 2050 maritime strategy underlines and strongly supports the role and growth potential of the UK port network in a post-Brexit world.  The British Port Authority’s launch of their autonomous shipping and ports network in August 2020 and the formation of the Mass Port’s Group, comprising of 8 nations (China, Finland, Denmark, the Netherlands, Korea, Japan, Norway and Singapore), in order to provide a thought-leadership think tank in relation to port automation and systems compatibility, shows the degree of attention which is being placed on the sector.

So what are the key drivers here?  As I have already said, the move to an “on demand” consumer requirement.  A closer connection between shipping and the major supply chain operators; corporations such as Amazon, Alibaba and Walmart and the significant potential that they may move closer and be more directly involved in the supply chain operation, not only running logistics and chartering vessels and boxes, but perhaps even moving to a full ownership operation. 

Then there are the savings in cost and the reduction of carbon emissions.  Enhanced logistical efficiency and the reduction of congestion.  Improved port planning and the maximisation of port space as a result of removing the human element.  Also, as part of the reduction of physical manpower operating in ports, the reduction of personal injury risk which itself can have a direct correlation with the efficiency of a port due to the inevitable delays which occur whilst incidents are dealt with and investigations carried out.  Finally, the elimination of supply chain blockages – never again would we need to fear that there will be no turkeys or Christmas trees for Christmas. 

(2) Non-Legal Challenges

But all of these potential gains are not without significant challenge. To truly take advantage of a universally automated supply chain we will need transparent data exchange, which is a challenge in an industry which still relies heavily upon completion and trade advantage.  Satoshi Nakamoto’s vision of a global blockchain has largely been frustrated due to the reluctance to openly share data.  While there has been considerable success in private blockchains, public blockchains and the vision of a system that could replace the international banking system, still seems far away.  And yet without such transparent data exchange will we ever be able to fully recognise the benefits of a fully automated supply chain? 

This indeed leads to another issue.  The possibility of systems between major ports, say, for example, the Ports of Singapore and London, having the ability to exchange data seamlessly may be solved relatively easily, but what of the world’s smaller and more remote ports? And how will funding for projects in more economically challenged jurisdictions be achieved? Indeed, the high cost and source of funding remains a major obstacle in port automation. 

Then there is the issue that advances in technology are moving far quicker than the required training and education of the specialist personnel that will still be required to update, maintain and supervise automated functions in ports. 

However, perhaps the single, largest challenge in an automated world is the social risk.  Reduction of the need for manpower in ports and throughout the global supply chain, vessel’s crews, crane operatives, truck drivers etc. will inevitably lead to mass unemployment.  We will need to address the social consequences that will likely follow.  Equally, we will have to consider how we deal with the loss of tax revenue as a result of a reduced workforce.  It seems likely that the source of tax revenue will have to move to the supply chain itself and therefore increase cost to the ultimate consumer. If this is the case, how much of the cost saving benefits will be eroded by a new taxation system?

Finally, and this next section could easily deserve a paper in its own right, there is the issue of cyber-risk.  We have seen a dramatic increase over the last three years in cyber-attacks on corporations and indeed on international ports.  Even more worrying is an increase in attacks focused on the maritime sector and operational technology in particular.  A number of counter-terrorism forums have identified the concern that the maritime sector may become increasingly targeted by cyber activists.  One only has to consider the dramatic impact that the grounding of the Ever Given in the Suez Canal had on the global supply chain to envisage how attractive to a terrorist, environmentalist or geo-political group such an attack may be.  Of course, there is nothing to suggest that what took place in relation to the Ever Given had any connection to cyber whatsoever. However, what is clear is that a hack of operational technology could easily have resulted in a similar casualty.   And the range of factors, geo-political, environmental, criminal and terrorist, itself presents a major challenge.  Without doubt we will have to see greater utilisation of cyber-protection and a widening of insurance protection were we to move to a fully automated supply chain.

(3) Legal Challenges

Everywhere we turn we seem to be faced by increased regulation.  In a recent research project carried out by Ince our clients identified that the number one most likely requirement of clients for legal services was in relation to guidance through increased regulatory compliance. 

Current port regulation will need to undergo significant and seismic review in order to make it fit for purpose. 

Another important consideration will be the issues that a port needs to bear in mind when comparing, for example, the hire of an employee as against the hire of a robotic service.  Issues of licencing, how we will identify breach, intellectual property rights, are all matters for consideration. Similarly the comparison between the leasing of a crane as against the leasing of the technology in order to automate a crane.  The legal challenges involved are far wider than in a traditional context. 

The same is true of claims and disputes.  In a traditional port context, legal issues generally track to two sources (i) fault of the machinery, (ii) fault of the operator.  However with automation that tracking matrix expands. We still have potential fault of machinery and fault of the remote operator, but added to this will be possible fault of the machine, defect in the AI, fault of the hardware, software, cyber-attack, loss of power, or the disruption of power supply as a result of a cyber-attack on major infrastructure. Another risk area is that of telecommunications failures.  The issue here being that the number of different technologies on which automation relies is far wider than in a traditional port context. 

Then there is a jurisdictional issue.  If remote operation is being conducted from a jurisdiction different to the location of the port, which law will apply? 

The exiting contractual framework will also need significant revision in order to address issues such as:

(i) Who owns the data?

(ii) Who is responsible for a failure of AI?

(iii) What legal recourse is available to a consumer or third party in the case of a failure of an automated system?

In a global on demand supply chain, where are the possible breaks in the chain of causation?  If one goes back to the Kalmar example, how will we assess liability in a true consumer on demand system?

Then there are issues surrounding the capture of personal data and preferences. We are already seeing challenges in relation to personal data collection.  In a more automated world these challenges are only likely to increase.  There will also likely be a need for a wholesale revision of health and safety and employment law regarding the interaction of humans and machines. 

We have already identified that with increased automation will come increased regulation, but if we over-regulate and restrict AI do we defeat its very purpose?  And finally, once again, the speed of technological advancement is greatly outpacing the development of law and regulation. 

I hope that over the last 25 minutes I have given you food for thought. Having started with a quote from ‘The Matrix’, I thought I should end with a quote from ‘The Terminator’.  I did think about John Connor’s line, “The future has not been written.  There is no fate but what we make for ourselves”.

But I settled with, “I’ll be back”.

Thank you.

Julian Clark is Global Senior Partner at legal and professional services firm, Ince. Julian is ranked in Chambers and Partners, the Legal 500 Hall of Fame, the US publication Super Lawyers and is a ‘Global Leader’ in ‘Who’s Who Legal’.

Project Leader, Dr Melis Özdel, is published in the Industrial Law Journal

The UCL Autonomous Shipping Project is pleased to announce that Project Leader, Dr Melis Özdel, has been published in the Industrial Law Journal this month (October 2021).

Dr Özdel’s article is titled: ‘Reconceptualising the Nautical Fault Exception in the Fog of Emerging Technologies’. The nautical fault exception removes, to some extent, sea carriers’ liability for losses arising from their employees’ acts or omissions. Dr Özdel considers how an adapted version of this exception should be available to govern the carriage of goods by new technology vessels. Specifically, she advocates assigning legal personhood to machine-learning algorithms (for the purposes of this exception) in order to provide efficient and fair risk allocation in the new era of shipping.

This exciting and topical article can be read in full through the following link:

https://academic.oup.com/ilj/advance-article-abstract/doi/10.1093/indlaw/dwab028/6383046

Interview with Felix Collin

In the third of a series of interviews involving experts in the fields of autonomous shipping and artificial intelligence, Project Leader Dr Melis Özdel and Project Editor Olivia Tolson speak to Felix Collin, doctoral candidate at the University of Turku. Felix is also a member of the Baltic Area Legal Studies steering group and the Algorithmic Agencies and Law research project. He recently contributed a chapter to the book Autonomous Ships and the Law (Routledge, 2020).

During the interview, Felix discusses aspects of his book chapter, including concerns surrounding the manning of autonomous ships, the potential basis for shipowner liability and the role of product liability in this context.

Anyone who wishes to engage with the discussion is encouraged to leave a comment below or on the YouTube video.

Interview with Paul Dean

In the second in a series of interviews involving experts in the fields of autonomous shipping and artificial intelligence, Project Leader Dr Melis Özdel and Project Editor Olivia Tolson speak to Paul Dean, Global Head of Shipping at HFW.

Lloyd’s List have recognised Paul as one of the top 100 most influential people in the maritime industry and one of the market’s top 10 lawyers in 2019 and 2020. Further, Paul contributes to HFW’s series of briefings on Autonomous Ships and recently co-wrote a chapter for the book New Technologies, Artificial Intelligence and Shipping Law in the 21st Century (2021). He also moderated the Second International Ship Autonomy and Sustainability Summit in 2020.

We very much enjoyed discussing with Paul how autonomous ships may impact the work of maritime law practitioners in the future, the cyber risk associated with their use, and when the right time to regulate autonomous shipping might be. His answers provide an excellent insight into this area. Anyone who wishes to engage with the discussion is encouraged to leave a comment below or on the YouTube video.

Interview with Professor Erik Røsæg

In the first of a series of exciting interviews with experts in the fields of autonomous shipping and artificial intelligence, Project Leader Dr Melis Özdel and Project Editor Olivia Tolson interview Professor Erik Røsæg of the University of Oslo.

It was an honour to speak to Professor Røsæg who has extensive experience in the area of Maritime Law. He recently edited, with Professors Henrik Ringbom and Trond Solvang, the book Autonomous Ships and the Law (Routledge, 2020).  This cutting-edge work explores autonomous shipping from a range of legal perspectives.

Following some introductory questions, our interview focuses on Professor Røsæg’s chapter from the book, ‘Diabolus ex machina’. His answers are thought-provoking and we would encourage anyone interested to contribute to the debate by leaving a comment on this blog post or on the video embedded below.

Welcome to the UCL Autonomous Shipping Blog

It is expected that the maritime world will be significantly different by 2050, from smart ports through to blockchain-based shipping documents and autonomous vessels.

The UK Government has recently published a report setting out its Maritime 2050 strategic ambitions, one of which is to ‘legislate for a domestic framework for autonomous vessels to attract international businesses’.

As the shipping industry is preparing for a sea change from labour intensive to autonomous shipping, it is timely to start a debate on the liability regime to be adopted for the carriage of goods by autonomous vessels.

The purpose of our research team is not to suggest an “ideal” liability regime, as the lessons from the past have shown that a widely accepted liability regime usually manifests itself as a compromise, leaving little room for considerations of ideal frameworks in purely legal terms.

In our quest for a solution, we proceed on the premise that our role as legal academics, lawyers and policy groups is an important one, since the traditional concepts of maritime law will have to be revisited together with our understanding of maritime operations and navigation.

It is hoped that this website will start an open and diverse debate, while also giving us a public engagement platform to disseminate our evidence-led findings to the wider audiences.