This week the theme is operations. After a season of announcements, the news is deployment: a giant port operator putting real money behind an AI platform across fifteen terminals, a major owner switching on real-time risk intelligence for its whole fleet, a tanker company betting all forty-four ships on voyage optimisation, and a cruise line handing its fuel buying to software. AI has moved from the pitch deck to the daily workflow — and a new security report is the reminder that once autonomous agents run your operations, they become part of your attack surface too. For managers, the takeaway is that the questions have shifted from "does it work?" to "what does it save, and what does it expose?"

The week's most important developments in shipping & oceans — distilled into a 5-minute read.

🔗 Quick Links

• 🇮🇳 Adani Ports (APSEZ) commits up to $100m to scale Kaleris AI across 15 terminals — Smart Maritime Network

  Multi-year deal deploys Kaleris' AI-augmented operating platform across 15 container terminals at nine ports, targeting +20% RTG crane productivity, +14% truck productivity and ~91 MMT of extra capacity by 2030

• 🛥️ MOL and IBM Japan switch on a generative-AI vessel-risk platform — Digital Ship

  Live since 1 July, the system built around MOL's Safety Operation Supporting Center uses generative AI to extract vessel risks in real time from weather, sea state, position and geopolitical data for 24/7 decisions

• 🧭 Uni-Tankers rolls Sofar Ocean's Wayfinder across all 44 tankers — Splash247

  The Danish tanker owner adopts fleet-wide voyage optimisation after platform users reported an average 6.9% fuel saving per voyage in 2025, tightening ship-to-shore routing and speed decisions

• 🧩 Veson Nautical folds its CoCaptain AI assistant into one unified platform — Veson Nautical

  The new Veson Platform brings IMOS, CoCaptain, AI-powered Mail and Bunker Insights into a single interface, with core CoCaptain included for every IMOS user and fuel-price predictions across 1,300+ ports

• 📦 Nexcade raises $6m to put AI agents into freight forwarding — The Loadstar

  Project A-led seed round backs AI agents already in live production with forwarders including XPO and Zencargo, plus a new "Atlas" tool that maps where work flows and where agents add the most value

• ⛽ Norwegian Cruise Line Holdings hands bunker procurement to ZeroNorth — Manifold Times

  Signed at Posidonia on 9 July, NCLH adopts ZeroNorth's Bunker Procurement Solution to pull procurement data, supplier engagement and purchasing workflows into one digital system

• 🔐 AI at large: autonomous agents now drive 1 in 8 AI breaches — HiddenLayer / PR Newswire

  As shipping wires agentic AI into the bridge and back office, a new threat report warns the average AI-agent breach costs ~$4.7m and only 14.4% of agents go live with full security sign-off

🚀 Big Moves This Week

1. Adani Ports Puts Up to $100m Behind AI — Across Fifteen Terminals

Adani Ports and Special Economic Zone (APSEZ), India's largest port operator, has expanded its partnership with US supply-chain software firm Kaleris into a multi-year, up-to-$100m programme that will deploy Kaleris' AI-augmented operating platform across 15 container terminals at nine domestic and international ports. This is not a pilot: it builds on first-phase deployments already live at six ports and scales advanced planning, optimisation and automation across the network. What the AI does: Kaleris' Advanced Optimization layer continuously recomputes how cranes, yard space and terminal trucks are sequenced, squeezing idle moves and dead time out of the operation. The numbers APSEZ is underwriting: the company expects up to a 20% improvement in Rubber-Tyred Gantry crane productivity and up to 14% in terminal-truck productivity, and projects the AI-led system will add roughly 91 million tonnes — about 10% of current capacity — by 2030. The spend sits inside a broader $850m outlay on decarbonisation, technology and capacity as APSEZ chases one billion tonnes of annual cargo handling. Why this matters for ship managers: the terminal operator increasingly owns the algorithm that sets your berth window and turnaround, and APSEZ handles a large share of India's container throughput — better crane and yard orchestration there flows straight into schedule reliability and demurrage exposure for lines calling those ports, and India already handles a large share of the country's container throughput, so improvements there ripple across an entire trade lane. The honest caveat: the headline gains are vendor-and-operator projections tied to a phased rollout through 2030, not audited results — so watch for measured, before-and-after turnaround numbers from named terminals as the deployment scales rather than taking the productivity percentages on faith. [Smart Maritime Network · APSEZ · Kaleris]

2. MOL and IBM Switch On Real-Time AI Risk Intelligence for the Whole Fleet

Mitsui O.S.K. Lines (MOL) and IBM Japan have launched a generative-AI platform, live since 1 July, that gives MOL's shore teams real-time risk intelligence to support 24/7 decisions across its global vessel operations. The system is built around MOL's Safety Operation Supporting Center (SOSC), the round-the-clock unit that watches over the fleet. The problem it targets: the data a watchkeeper needs — weather, sea state, vessel operating data, navigation status and geopolitical information — has historically lived in separate systems, forcing slow manual stitching in exactly the moments when speed matters. What the AI adds: the platform uses generative AI to identify and extract vessel risks in real time, drawing on both historical operating records and current data, so the SOSC can spot an emerging problem and act on it faster than a human trawling multiple screens. It is framed squarely as decision-support for a crewed, human-led operation, not autonomy. Why this matters for ship managers: rising navigational and geopolitical risk — congestion, weather extremes, conflict-adjacent routing — is making real-time situational awareness a competitive capability, and MOL is one of the largest operators in the world putting a named system into daily use rather than a lab, at a moment when a single missed risk can cascade into a costly diversion or insurance claim. It is a strong "deployed, not demonstrated" marker for AI in fleet safety and operations. The honest caveat: MOL and IBM have not yet published outcome metrics — incidents avoided, decision-time saved, or deviations reduced — so the real test is whether the SOSC can show that faster risk extraction changed decisions and results, not just consolidated more data onto one screen. [Digital Ship · Marine Insight]

3. Uni-Tankers Bets All 44 Ships on Voyage Optimisation

Danish tanker owner Uni-Tankers has signed a partnership with Sofar Ocean to deploy its Wayfinder voyage-optimisation platform across the company's entire 44-vessel fleet, which trades in both the spot market and on long-term contracts. The move is a full fleet-wide commitment rather than a trial on a handful of ships. What Wayfinder does: it turns weather, sea state and vessel-performance data into route and speed recommendations, and — crucially — tightens the coordination between the ship and the shore team so the whole operation is working from the same optimised plan rather than second-guessing each other. The number that anchors the decision: Sofar Ocean says shipowners using the platform in 2025 achieved an average fuel saving of 6.9% per voyage — a figure that, applied across a 44-ship tanker fleet, translates into meaningful fuel cost and emissions reductions and directly supports CII and FuelEU compliance. Uni-Tankers frames the deal as part of a strategy pairing operational efficiency and digital capability with fleet growth. Why this matters for ship managers: voyage optimisation is now one of the clearest, lowest-risk AI wins available — no hardware refit, measurable savings, and a compliance dividend — which is exactly why a growing operator is willing to standardise on it across every hull rather than run it on a cautious handful of ships first. The honest caveat: the 6.9% figure is a platform-wide average reported by the vendor, not a Uni-Tankers result, and real savings depend on trade pattern, baseline discipline and how faithfully crews follow the recommendations — so the number to watch is Uni-Tankers' own before-and-after fuel performance once the fleet is fully live. [Splash247 · Smart Maritime Network · Digital Ship]

4. Veson Folds Its AI Assistant Into One Platform — the Consolidation Bet Again

Veson Nautical, whose IMOS software is the operational backbone for many commercial shipping companies, has launched the Veson Platform, a unified environment that brings its flagship IMOS together with the CoCaptain AI assistant, AI-powered Mail, and new Vessel Insights and Bunker Insights tools under a single interface. The pitch is that rather than bolting AI on as a separate app, Veson embeds it into software owners, operators and charterers already run every day. What's included: core CoCaptain is now bundled for all IMOS users at launch, letting them query contracts, voyages, communications and market intelligence in one place; Bunker Insights is a machine-learning tool trained on hundreds of thousands of bunker transactions to generate daily fuel-price predictions across more than 1,300 ports. Why this matters for ship managers: this is the consolidation-versus-best-of-breed debate made concrete again — the appeal of a single, AI-infused system of record is fewer integrations and cleaner data, and bundling the AI assistant removes the "do we buy the copilot?" decision for existing customers, quietly making AI the default rather than an upgrade they must justify. It directly tests our standing question about whether back-office AI copilots deliver audited ROI, not just time saved. The honest caveat: an incumbent bundling AI into a platform many operators already depend on raises the single-vendor lock-in question the market keeps flagging — the value will hinge on whether CoCaptain produces decisions users trust, and whether Bunker Insights' price predictions prove accurate enough to buy on. Ask for measured outcomes and an honest read on data portability before deepening the dependency. [Veson Nautical · MarineLink]

5. Nexcade Raises $6m to Put AI Agents Into Freight Forwarding

Nexcade, a startup building AI agents for freight forwarders, has raised a $6m seed round led by Project A Ventures, with Connect Ventures, MMC Ventures and Entropy Industrial Capital also participating. Since emerging from stealth in October 2025, the company has moved from early design partnerships into live production deployments with forwarders including XPO, Zencargo, Cardinal Global Logistics and CargoTrans across Europe and the US. What the agents do: they automate the document-heavy, communication-heavy back office of forwarding — the quoting, coordination and data entry that eats forwarder time — executing multi-step tasks rather than just flagging them. Alongside the raise, Nexcade is launching Atlas, a data-insights tool that analyses a forwarder's communication data to show where teams spend time, where processes break down, and where AI agents can have the largest commercial effect — a way to target automation at the highest-value bottlenecks first. Why this matters for ship managers: freight forwarding is the connective tissue of the supply chains that shipping serves, and agentic AI landing in live forwarder operations — with named customers, not logos on a slide — is a signal that "digital workers" are becoming real operational tooling in logistics. It follows Cargofy's raise last month, marking freight-AI funding as a genuine trend rather than a one-off, and hinting that investors now see forwarders' back offices as fertile ground for automation. The honest caveat: this is a seed round, and the durable test is retention — whether XPO, Zencargo and the others expand the agents across more workflows and keep them in production, or quietly wind the pilots down once the novelty fades. [The Loadstar]

6. Norwegian Cruise Line Hands Bunker Procurement to ZeroNorth

Norwegian Cruise Line Holdings (NCLH) has adopted ZeroNorth's Bunker Procurement Solution to modernise how it buys fuel, in a partnership signed at Posidonia 2026 in Athens on 9 July. For a cruise operator, fuel is one of the largest and most visible operating costs, and the deal moves procurement out of scattered spreadsheets, phone calls and email chains into a single digital workflow. What the platform does: it brings procurement data, supplier engagement and purchasing workflows into one system, giving the buying team a structured, transparent view designed to support faster, data-driven decisions and tighter collaboration with suppliers. In a market where the price achieved on any given stem has traditionally been hard to benchmark, digitising the workflow introduces the price discovery and audit trail that manual buying lacks. Why this matters for ship managers: bunker procurement is one of shipping's oldest and most opaque cost centres, and a major branded owner putting its fuel buying onto a digital platform continues the shift — visible in Asyad's AuctionConnect commitment and others — from bilateral, relationship-driven purchasing toward structured, data-backed procurement. The question it poses to every fleet is whether their current bunker process would survive comparison with a system built to make every stem visible and benchmarked. The honest caveat: the announcement describes workflow modernisation, not yet a reported saving — the value of digital procurement is proven in dollars-per-tonne and reduced disputes over time, so the metric to watch is whether NCLH or ZeroNorth publishes measurable procurement savings once the system beds in, and whether the promised transparency translates into a better price on the stem rather than just a tidier paper trail. [Manifold Times · Ship & Bunker]

7. As Shipping Goes Agentic, the Security Bill Arrives — 1 in 8 AI Breaches Now Come From Agents

A new report from AI-security firm HiddenLayer warns that autonomous "agentic" AI has become a leading enterprise attack surface — and shipping, which spent the past year wiring agents into the bridge and back office, is squarely in the blast radius. We flag this as our wider-AI story because the same capability driving this year's maritime deployments is the capability that creates the risk. The findings: HiddenLayer's 2026 AI Threat Landscape Report says autonomous agents now account for more than 1 in 8 reported AI breaches, the average AI-agent breach costs about $4.7m, and 48% of security professionals rank agentic AI as their top attack vector for 2026 — ahead of deepfakes. Most striking for anyone deploying fast: only 14.4% of AI agents go live with full security and IT approval, 76% of organisations cite "shadow AI" as a growing problem, and 31% do not even know whether they suffered an AI breach in the past year. The maritime connection: agents that can browse data, call tools and execute multi-step workflows are exactly what platforms like MOL's risk system, Veson's CoCaptain, Ofiniti's MCP connectors and freight agents such as Nexcade's now put into daily operations — expanding the attack surface across navigation, chartering and procurement. Why this matters for ship managers: this is the counterweight to every deployment story above. The takeaway: treat AI security and data integrity as a first-class line in procurement — ask vendors how agents are sandboxed, approved and audited — before an unmonitored copilot becomes an unmonitored incident. The report's lesson is blunt: the speed that makes agentic AI valuable is the same speed that outruns the controls meant to govern it. [PR Newswire · HiddenLayer]

📊 Why It Matters — Strategic Impact Table

🔭 On Our Radar

• 🇮🇳 Do AI Port Platforms Finally Publish Hard Turnaround Numbers? — Adani's up-to-$100m Kaleris rollout follows China Merchants Port's smart-logistics push, we monitor whether either operator publishes measured turnaround, crane-productivity and yard-efficiency gains from named terminals, track whether AI berth and crane orchestration becomes standard infrastructure at major hubs, and assess what terminal-side control of the berth window means for lines trying to plan just-in-time arrival.

• 🛥️ Does Real-Time AI Risk Intelligence Change Decisions, or Just Consolidate Screens? — MOL and IBM have put generative-AI risk extraction into a live 24/7 operations centre, we monitor whether MOL publishes outcomes — incidents avoided, deviations reduced, decision-time saved — track whether other large operators build equivalent real-time risk platforms, and assess whether "situational awareness AI" proves its value beyond bringing more data onto one screen.

• 🧭 Do Voyage-Optimisation Fuel Savings Hold Up Fleet-Wide? — Uni-Tankers has bet all 44 tankers on Wayfinder on the strength of a 6.9% platform-average saving, we monitor whether the owner reports its own before-and-after fuel numbers, track how much of the modelled saving survives real trade patterns and crew adherence, and assess whether fleet-wide voyage optimisation becomes the default compliance-and-cost play under CII and FuelEU.

• 🧩 One Platform or an Ecosystem — and What About Lock-In? — Veson bundling CoCaptain into IMOS deepens the single-system-of-record model, we monitor whether owners favour integrated AI-infused suites or interoperable best-of-breed stacks, track whether bundled copilots show audited ROI rather than time-saved claims, and assess how managers preserve data portability and avoid lock-in while still getting a connected operational picture.

• 📦 Does Freight-Forwarding AI Funding Convert to Retained Deployments? — Nexcade's $6m seed follows Cargofy's raise, marking a fundable freight-AI-agent trend, we monitor whether named forwarders like XPO and Zencargo expand agents across more workflows or quietly retire the pilots, track whether "digital workers" show durable productivity gains, and assess how agentic logistics tooling reshapes the supply chains shipping serves.

• ⛽ Does Digital Bunker Procurement Produce Reportable Savings? — NCLH's move onto ZeroNorth continues the shift from relationship-driven to structured fuel buying seen with Asyad's AuctionConnect deal, we monitor whether owners publish measurable dollars-per-tonne savings and fewer disputes, track how many major operators digitise procurement over 2026, and assess whether bunker platforms integrate with voyage optimisation and CII tools into end-to-end fuel management.

• 🔐 As Maritime Goes Agentic, Does Cyber-Readiness Keep Pace? — HiddenLayer reports autonomous agents now drive more than 1 in 8 AI breaches while only 14.4% go live with full security sign-off, we monitor whether operators treat AI security and resilience as a first-class requirement, track how vendors evidence sandboxing, approval and audit of AI agents, and assess whether "shadow AI" gets inventoried before an unmonitored copilot becomes an incident.

📅 Critical Maritime AI Research Areas for Managers

1. 🇮🇳 Measured Turnaround Outcomes From AI Terminal Platforms: Adani's Kaleris rollout and China Merchants Port both promise crane, yard and berth gains — research should compare projected versus achieved productivity and turnaround at named terminals, giving lines an evidence base for how much AI port orchestration actually improves schedule reliability and demurrage exposure rather than vendor percentages.
2. 🧭 Realised Versus Advertised Voyage-Optimisation Savings: Platform-average fuel-saving claims like Sofar Ocean's 6.9% need grounding in operator reality — research should track achieved fuel and emissions outcomes across fleets, trade patterns and levels of crew adherence, so owners can size the CII and FuelEU dividend of fleet-wide voyage optimisation before committing every hull.
3. 🛥️ Decision-Impact Metrics for Real-Time Risk-Intelligence Systems: MOL and IBM's SOSC platform bets that faster risk extraction improves 24/7 decisions — research should define repeatable measures (incidents avoided, deviations reduced, decision latency) so operations centres can judge whether AI situational awareness changes outcomes, not just how much data it consolidates onto one screen.
4. 🔐 AI Security and Agent Governance for Maritime Operations: With autonomous agents now a leading breach vector, research should establish maritime-specific guidance for sandboxing, approving and auditing AI agents across navigation, chartering and procurement, plus practical ways to detect shadow AI — turning "AI cyber-readiness" from a slogan into a checklist owners can apply during procurement.
5. 📐 Deployment-Maturity Benchmarks for Maritime AI Claims: The "deployed versus demonstrated" gap runs through MoUs, product launches, funding rounds and adoption figures alike — research should maintain a simple, repeatable maturity benchmark (named operators, fleet size, deployment length, published metrics) that ship managers can apply during vendor evaluation, turning announcements of every kind into comparable, evidence-based decisions.

📈 Top Investment Opportunities

1. 🇮🇳 AI Terminal Orchestration Software — Adani's up-to-$100m Kaleris programme shows major operators treating AI berth, crane and yard planning as core terminal infrastructure — the investment opportunity is in optimisation platforms that turn unpredictable turnarounds into schedulable, benchmarked performance and can prove measured gains across ports rather than locking a terminal into one stack — Adani + Kaleris
2. 🧭 Voyage-Optimisation Platforms With a Compliance Dividend — Uni-Tankers standardising Wayfinder across 44 tankers shows real pull for no-hardware fuel-saving software that doubles as a CII and FuelEU tool — the investment opportunity is in voyage-optimisation and weather-routing platforms that can demonstrate audited, operator-specific savings and tie routing decisions to emissions compliance — Uni-Tankers + Sofar Ocean
3. 🛥️ Real-Time Maritime Risk-Intelligence Systems — MOL and IBM putting generative-AI risk extraction into a live operations centre signals demand for situational-awareness tooling as navigational and geopolitical risk rises — the investment opportunity is in platforms that fuse weather, position, performance and geopolitical data into real-time, decision-grade risk intelligence for shore teams — MOL + IBM
4. ⛽ Digital Bunker Procurement Platforms — NCLH moving fuel buying onto ZeroNorth continues the shift from opaque, relationship-driven purchasing to structured digital procurement — the investment opportunity is in platforms that bring price discovery, supplier engagement and audit trails to bunkering, especially those integrating with voyage optimisation and CII tools into end-to-end fuel management — NCLH + ZeroNorth
5. 🔐 Maritime AI Security and Agent Governance — HiddenLayer's data shows autonomous agents becoming a leading breach vector just as shipping adopts them — the investment opportunity is in the security, sandboxing, monitoring and audit layer for AI agents, the essential counterpart to every maritime AI deployment and a requirement that will move from optional to mandatory in procurement — HiddenLayer report

📅 Top Monthly Picks

1. ⛽ Ofiniti Buys Teqplay — a Bunkering Platform Reaches for the Whole Operational Picture — the DNV-spun-out fuel-operations platform acquired Rotterdam digital-twin specialist Teqplay to fold real-time port-call intelligence into bunker operations, the clearest sign yet that integrated operations platforms are absorbing the standalone tools around them — Ofiniti + Teqplay
2. 🏛️ Korean Register and Microsoft Put AI Agents Into Ship Classification — KR's cooperation with Microsoft Korea embeds AI agents across classification, inspection and technical services and plans an AI-centred ship data centre, a sign that the regulatory trust layer itself is going digital and reaching every vessel on a fixed cadence — KR + Microsoft
3. 🚢 Hafnia Says Its Enterprise-AI Rollout Is Already Paying Off — the major tanker owner expanded its Complexio AI deployment across the business and reported operational gains, a rare and valuable example of a large operator treating enterprise AI as a working discipline with results, not another pilot — Hafnia + Complexio
4. 🏗️ NC AI and Hanwha Ocean Go After the Hardest Job in the Yard — Autonomous Welding — NC AI won a Hanwha Ocean contract to build an autonomous physical-AI welding system for shipyards, pushing computer vision into one of the most hostile environments it can face and testing whether shipyard automation eases the capacity pressures that reach owners as delays — NC AI + Hanwha Ocean
5. 🤖 Maritime Robotics Banks Fresh Capital to Scale Unmanned Vessels — the Norwegian firm raised a substantial round to scale uncrewed-vessel technology for commercial and defence customers, a marker of continued capital flowing into maritime autonomy — the open question is which of these investments convert into named commercial operators with published results — Maritime Robotics

Have feedback or a story tip? Email us or connect on LinkedIn.