Trucking Firms Face AI-Enabled Fraud Schemes

Trucking and logistics companies face a surge in AI-driven fraud. (THAWEEKIET SRIRING/ Getty Images)

As trucking and logistics companies invest in cybersecurity measures, fraudsters are advancing just as quickly, using the latest generative and agentic artificial intelligence to refine schemes and scale them across the industry. The result is an increasingly fast-moving contest in which both sides are learning and adapting in real time, industry veterans and technology experts said.

“The tricky thing about AI is we have a great set of tools to advance fraud detection and prevention, but the bad actors have the same tools to create even more convincing techniques,” said David Mitsui, director of risk and compliance operations for freight broker Total Quality Logistics.

Generative AI creates realistic content, while agentic AI can go further by acting like employees and executing tasks. In practice, that has translated into more sophisticated impersonation attempts.

TQL, which ranks No. 9 on the Transport Topics Top 100 list of the largest logistics providers in North America, has seen an uptick in falsified shipment documents, altered insurance certificates, manipulated driver’s licenses and edited images of equipment and cargo.

“These aren’t one-off attempts. They’re coordinated, continuous and constantly adapting,” Mitsui said.

The nature of fraud in trucking has fundamentally shifted. What was once largely straightforward cargo theft has evolved into identity-driven, AI-enabled impersonation and credential abuse that can be deployed at scale.

The wider availability of AI tools is enabling criminals to exploit verification gaps, assume trusted identities and move freight under false pretenses. Industry participants are working to understand how these tactics are being used and to build layered defenses to counter them.

Ben Barnes, chief information officer for McLeod Software, said the industry is seeing both an increase in the speed of attacks and a lower barrier to entry for actors who previously would not have had the technical capability to execute complex schemes.

The industry’s reliance on relationships and long-standing trust compounds the risk. Many transactions still move quickly based on assumed familiarity, creating openings that can be exploited when identities are convincingly mimicked.

Data from Verisk CargoNet points to the same trend. Its first-quarter 2026 supply chain risk analysis described impersonation-based theft as a systematic and scalable criminal method. While overall crime events declined compared with the same period a year earlier, the report emphasized that organized networks are continuing to refine their tactics.

Those networks frequently rely on credential harvesting to gain access to transportation systems. Phishing campaigns and malware are used to compromise email accounts, internet-based phone systems and load-matching platforms. With those credentials, attackers can pose as legitimate carriers, accept tenders, communicate with brokers and ultimately redirect freight.

“We’re watching transnational organized crime groups become the dominant force in the cargo theft landscape,” said Keith Lewis, vice president of operations for Verisk CargoNet. “There is a clear preference for goods that move easily through online resale channels.”

Generative AI is making those schemes more precise. Chris Burroughs, president of the Transportation Intermediaries Association, said the technology enables highly convincing communications and identities at a scale that was not possible just a few years ago.

“In an industry built on speed and communication, this creates real operational risk for the entire industry,” he said.

Image

Johnson

Ryan Johnson, director of cybersecurity for Trimble, said attacks are increasingly crafted within operational context, incorporating internal roles, timelines and industry-specific language to make them more believable. That context-aware approach makes it more difficult for employees to distinguish legitimate interactions from fraudulent ones.

As companies adopt their own AI tools, they face an additional layer of complexity. Agentic systems must be managed carefully, with clearly defined access and constraints.

“Establishing solid governance via an AI policy and working board is always a good start,” Johnson said.

Ben Wilkens, director of cybersecurity for the National Motor Freight Traffic Association, compared agentic AI tools to an overeager intern with too much access. They are capable of executing assigned tasks, but they do so without full awareness of unintended consequences.

“They will operate in a way to achieve those outcomes, but they’ll do so blindly,” he said.

That dynamic underscores the need for tighter controls in a complex operating environment. Trucking and logistics companies must determine what an agentic system should be able to access and ensure guardrails are in place so it can operate effectively without introducing additional risk.

At the same time, defenders face a persistent structural challenge. Attackers need only find a single weakness, while companies must secure multiple systems and processes across their operations.

A layered approach to security has emerged as standard practice. Individual controls may not be sufficient on their own, but combined measures can limit the pathways available to an attacker.

AI Cyber Defenses

Technology providers are responding by embedding AI-driven validation tools within their platforms. These systems are trained on datasets of known fraudulent documents and behaviors, allowing them to identify anomalies and provide confidence scores that help operators assess risk before taking action.

“The model identifies patterns and returns a confidence score you can make assertions against,” said Adam Masterson, senior director of application development for PCS Software. “It’s not a guarantee, but it gives companies a data-backed signal to act on.”

 

Image

 

The AI Summit at TMC

The Technology & Maintenance Council and Transport Topics will co-host a new event at TMC’s 2026 Fall Meeting, Sept. 20-24, to advance the conversation about artificial intelligence implementation in the trucking industry. Read more

At PCS Software, those capabilities are integrated into transportation management systems to support dispatchers rather than replace them. The intent is to provide better information at the point of decision-making.

Other companies are applying AI to monitor activity in real time, flag inconsistencies during carrier onboarding and surface potential risks before a shipment moves. TQL’s Mitsui said such tools can enhance awareness but must be combined with disciplined processes.

“The technology is getting better, but it has to be paired with the right processes and people to be effective,” he said.

Companies are reinforcing those processes through stronger onboarding protocols, multi-step verification and continuous monitoring of carrier and customer activity. Human judgment remains a central element, particularly in situations that fall outside normal patterns.

Prevention efforts are also extending into automation. Trimble’s Johnson pointed to emerging platforms focused on security orchestration, automation and response, which can isolate compromised systems, revoke access and address vulnerabilities with minimal delay.

Verification and Training

Even as technology evolves, basic verification practices remain essential.

Companies are investing more heavily in multi-step verification processes, stronger onboarding protocols, internal escalation procedures, employee training and continuous monitoring of carrier and customer activity, TIA’s Burroughs said.

“Human judgment still matters enormously, especially when something feels inconsistent or unusual,” he added.

Lewis of Verisk CargoNet said companies should routinely recheck identities and scrutinize communications that fall outside standard workflows.

“Ask the person on the other end to drop their VPN,” he said, noting that doing so can help verify location.

Operational discipline is also critical. Masterson of PCS said companies should identify common scenarios that could be exploited and ensure employees understand how to handle them. Requests that arrive through unfamiliar channels or deviate from established processes should be treated with caution.

McLeod’s Barnes emphasized that trust must be earned at each step of a transaction.

“If you get an email with somebody’s name and number on it, don’t call that number,” he said. “Call the number you know.”

Image

Mitsui

Training has become more widely distributed across organizations as a result. Mitsui said awareness of fraud risks must extend beyond IT teams to include operations, sales and compliance.

“Training can’t just live with the IT team; operations, sales and compliance all need to understand the risks,” he said. “The companies that get this right will treat AI as a tool to enhance their people, not replace the judgment that keeps the business safe.”

Industry attitudes toward fraud are also evolving. Barnes said there was once a reluctance to discuss incidents publicly because of reputational concerns. That has shifted as companies recognize that cybercrime groups are targeting the sector broadly.

“Five or six years ago, you didn’t talk about it because it was damaging your brand,” he said.

Increasingly, companies are acknowledging that the problem cannot be addressed in isolation. As fraud tactics become more coordinated and technologically advanced, collaboration across carriers, brokers, shippers, technology providers and law enforcement is becoming more important.

“We’ve got to fight the fraud as one,” Barnes said. “Or else we’re just hit with one-offs all over the place.”