From the streets of Bangkok to the future of terminal safety, this week’s newsletter brings together a few reflections on AI, people, safety and the realities of implementation.
From CTAC Asia 2026 and conversations around AI and automation, to a bigger question:
What should AI do — and what should it never decide?
Also, a look at how Qavach is moving from analysing incidents to helping teams anticipate and act on risk.
A few thoughts, a few lessons, and a lot to reflect on.
Siam !! Bangkok !!
There is never a time the city fails to impress and to send every visitor back with a treasured memory.
This one was about how well the 22 million city cohabit; the joy of a Grab motorbike taxi; a ride on the Orange River taxi in a huge storm; about the most amazing food; visit to the flower market at 5 A.M; prayers at the Erawan Shrine.
It was about the people; the hospitality; the discipline; the calm; the energy……………..just endless things.
And it was about the Container Terminal Automation Conference, where Quantum and I were panellists speaking about AI as well as Safety. (more on that next week).
It was about meeting some great people who attended the conference (more next week) as well as fellow panellists (also - more next week).
I met old friends; people who knew old friends; people with whom we shared great stories; and people who spoke about the many issues that affect the industry and solutions we need (More next week 😊 )
And finally, a call out to the team who put this together - Benjamin Heather and Inês Santos . Thank you for an amazing job and having us in Bangkok.
CTAC Asia 2026 Bangkok made one thing clear: the conversation around AI and automation is moving beyond enthusiasm and into execution.
It was a pleasure to share the stage with Abby Z. , Sangkom K. , and Alex Van Winckel during the session Artificial Intelligence in the Terminal: From Pilots to Productivity. And Anat Machima Riyaz Melekolangath Iresh Siriwardena MBA - PIM (Uni of Sri Jayawardenepura) during Safety as a Performance Lever
One point I emphasised during the discussion was simple: AI cannot be measured by how sophisticated it sounds. It has to be measured by what it delivers.
For terminal operators, the question should not be “Do we need AI?” but: “What problem are we solving, and what measurable improvement will AI create?”
Every AI investment needs a clear connection between cost, performance and outcomes. For me, the value of AI in terminals comes down to three fundamentals:
Visibility. Productivity. Accountability.
AI should give operators better visibility into what is happening, help teams become more productive, and create greater accountability around decisions and outcomes.
The discussion also highlighted that AI adoption is not simply about introducing another technology layer. Terminals already generate enormous amounts of data through their operational systems. The opportunity is to connect that data, make sense of it and turn it into intelligence that people can act upon.
And this is where the industry needs to be careful. We should not automate simply because we can. We should automate where it creates measurable value.
That is exactly where the next generation of terminal intelligence will be tested.
👉🏼 Read more on Port Technology International
The more we learn about AI, the more we realise how much we still have to learn.
For terminal safety– as for many sectors– this is both exciting and challenging.
At Quantum, we have built systems that automate, integrate and help predict risk. But we also believe true intelligence goes further– recognising patterns, learning from data and turning insights into prescriptive action.
We would like you to imagine a safety system that can:
This is where the future of safety is heading.
In this video from some months ago, I spoke about how Qavach was developing, helping us move from analysis after the fact, to intervention before it happens. With Qavach taking significant strides, it is good to see how our principle and approach to Artificial Intelligence remain grounded and proactive.
I firmly believe the goal should not be AI for the sake of AI. It The aim is intelligence that helps people make better decisions, earlier.
▶️ Watch: The Next Frontier – AI and the Future of Safety
The AI conversation in our industry often focuses on what technology is able to pull off, incremental new things it can do. Detect! Predict! Automate!
An equally important question though? What should AI never decide on its own? With Qavach, we have deliberately drawn that line.
AI can detect entry into a restricted zone, identify unusual behaviour, rank emerging risks- but the decisions that carry human consequences– stopping an operation, sending someone to investigate, deciding immediate action– these remain with people.
That is not a limitation. It is a design choice.
AI should handle the relentless work of watching, analysing, finding patterns. Humans should retain the judgement.
👉🏼 Read more on LinkedIn
