In 2024, an employee at global engineering firm Arup joined a video call with the company’s chief financial officer and several colleagues—or so it seemed.
The execs on screen looked and sounded real. Only they weren’t. Fraudsters had used deepfake technology to impersonate them. By the time the employee realized the deception, they had made 15 transfers worth roughly $25 million. Arup later confirmed that fake voices and images were used in the attack.
Similar deepfake tactics are working their way into supply chains. Instead of asking someone to transfer money, a fraudster might ask them to reroute a shipment, change a delivery address, release freight to a different carrier, or share sensitive information.
When a request looks and sounds legitimate, people are more likely to trust it. And that’s exactly what fraudsters are counting on.
How deepfakes could show up in logistics
You may have first heard about deepfakes through an AI-generated celebrity video, political clip, or cloned song. At their core, deepfakes are synthetic or digitally manipulated videos, audio, or images designed to imitate a real person.
But impersonation and shipment diversion aren’t new problems. We’ve previously written about how cargo theft in the digital age can begin with a phishing email, a spoofed website, a fraudulent phone call, or someone posing as a legitimate freight hauler.
What’s changed is that AI is now making these scams much harder to detect.
Here’s how that could play out in everyday logistics operations:
A last-minute request to reroute a shipment
An employee receives a message from someone who appears to be an authorized customer contact. There’s been a change of plans, they explain, and the shipment needs to go to a new address. A follow-up call—even a video call—appears to confirm the request.
On paper, everything may look right. But if no one independently verifies the change, the shipment could be headed straight to a criminal-controlled location.
A fake carrier or driver confirmation
A fraudster poses as a dispatcher, carrier representative, or driver who needs to collect or redirect a load. AI-generated audio or video could help that person match a known contact or make falsified credentials seem more credible.
An urgent request from a senior leader
An employee gets a call that sounds like an executive asking them to skip a standard approval step to keep a high-priority shipment moving. The explanation may sound reasonable. Logistics teams solve urgent problems every day.
That’s where things get tricky. A request designed to bypass normal controls can look a lot like good customer service in the moment.
A request for sensitive information
Someone posing as a customer, carrier, supplier, or colleague asks for shipment details, customer data, or login information. Even if the first request doesn’t lead directly to stolen freight, the information can help build a more convincing scam later.
These examples are hypothetical, but the tactics behind them are already familiar. Fraudsters exploit trust, incomplete information, and pressure to act quickly. Deepfakes give them another way to make the story hold together.
Why logistics is a natural target for deepfake scams
Supply chains depend on fast communication among customers, carriers, drivers, warehouse teams, vendors, and logistics providers. A single shipment may move through several systems, locations, and organizations before it reaches its destination.
Plans change. Exceptions happen. Teams communicate by email, phone, text, and video—and delays can get expensive quickly.
That makes logistics especially vulnerable to a well-timed impersonation. An unexpected request may not seem all that unusual when handling exceptions is part of the job. And when a load is waiting, there’s pressure to solve the problem now.
That sense of urgency keeps goods moving. It can also make people more likely to act before they’ve verified who’s asking.
One false request can cause a much bigger problem
The average stolen-load value is approximately $214,000. But the cost of a diverted or stolen shipment rarely stops with the goods themselves.
A company may also have to cover:
- Investigation, recovery, and remediation
- Delayed or missed deliveries
- Inventory shortages and fulfillment disruptions
- Replacement goods and transportation
- Employee time spent responding to the incident
- Contractual, insurance, privacy, or compliance issues
Then there’s trust.
Customers need to know that only authorized people can change delivery instructions or access sensitive information. Carriers and logistics providers need confidence that the people and businesses involved in a load are legitimate.
One successful impersonation can cast doubt on other communications, strain partner relationships, and raise uncomfortable questions about the controls that failed. If the incident becomes public, the reputational damage may last much longer than the operational disruption.
How Kenco helps protect its customers
At Kenco, our goal is to help our customers keep their supply chains resilient. Part of that effort is to protect them from new digital risks, like deepfakes.
Our approach doesn’t rely on employees being able to identify every deepfake. Instead, we build safeguards into everyday operations so that one convincing (but unverified) message can’t easily change the flow of goods.
Verifying consequential changes
Requests to change shipment details or delivery addresses go through multiple verification steps. An email, phone call, or video interaction alone isn’t enough to approve the change.
Escalating unusual requests
Employees have clear paths for raising requests that seem unexpected, inconsistent, or out of the ordinary. Urgency may be part of the job, but it isn’t a reason to skip verification.
Training teams to recognize the warning signs
Kenco trains employees to recognize suspicious communications and AI-generated anomalies, question unexpected requests, and follow the right process when something doesn’t add up.
Protecting sensitive data
Deepfake scams become more convincing when criminals have accurate information about a company, its employees, or its shipments. Kenco’s data governance and compliance practices help protect customer information and reduce the chance that it will be accessed or misused.
These safeguards are part of Kenco’s broader security and risk-management work. In 2025, Kenco IT analyzed 23.8 million emails, flagged nearly one million security events, and remediated thousands of attacks. Proactive monitoring and security tools helped contain threats before they disrupted operations or damaged customer trust.
Kenco also has dedicated compliance and risk teams that vet and monitor transportation providers and support customers during active fraud or theft incidents. No single control does the whole job. Together, technology, training, operational checks, and strong governance make it much harder for a false request to become a real disruption.
Trust—but back it up with a process
Deepfakes will keep getting better. It will become even harder to tell whether a familiar face, voice, or message is real based on appearance alone. Companies need processes that stop unverified information from turning into action.
However, that doesn’t mean treating every customer request with suspicion. It means slowing down at the moments that matter: a new address, a different carrier, an unexpected request for information, or pressure to bypass a standard check.
AI will continue to change how fraud looks, which means we’ll continue strengthening our safeguards to stay one step ahead