Here’s How AI Agents Can Protect EV Chargers

Number of electric vehicles streets around the world continue to grow. The increasing adoption of EVs has led to the development of more accessible, faster, and better control systems.

However, this expansion also introduces new cyber security risks that have not been widely studied, and for which there are few possible solutions.

Cristina Alcaraz, a structural safety researcher at the University of Malaga in Spain, explains that the role of charging an electric car stations because they combine several physical and digital components. It is said that this complex construction not only hinders the performance of chargers but also poses many safety issues. Exposure to charger attacks disrupts the continued adoption of EVs and destabilizes power grids in countries where chargers operate.

With the aim of combating this threat, researchers from the NICS laboratory at the University of Malaga have developed a new concept for deployment. AI assistants to protect infrastructure. These agents are designed to prevent cyberattacks from a variety of vectors, from fraud or electronic theft by malicious actors using toll booths to large-scale attacks that can destroy the most powerful networks.

The group’s idea is to ensure early and reliable detection of errors and attacks in the networks they are using Open Charge Point Protocol. The OCCP standard is one of the most widely used in the management and control of electric chargers. The protocol allows a network of payment centers to communicate with a central system that can monitor, manage, and coordinate all electronic services used by end users.

The central system takes care of a bunch of things remotely, including user authentication, power flow management at each station, power usage monitoring, and technology monitoring. This capability allows for real-time control and enables users to see and respond quickly to any unusual behavior.

However, the authors of the new research show that the current methods of analysis that are being used using this process are often focused on the Internet or local events, so they can only give a limited idea of ​​what is happening in the entire construction area. The researchers say this limitation makes it difficult to determine where system errors are occurring, which network devices are compromised, the extent of each threat, and the ways in which threats can spread.

Make a phone call in AI

The researchers propose a system that uses multiple AIs. Each train or key part of the charging network includes AI agents that can analyze their environment, gather information, and collaborate with other agents to better visualize the environment.

“Each service provider checks the status of chargers, connections, and connected devices to detect malfunctions, performance failures, or security incidents,” says Alcaraz. “These agents, linked to a central control system, compare local data with those of nearby stations, providing a complete, accurate, and coordinated view of the situation,” he says. Alcaraz is the same the main report writer.

The task, printed in International Journal of Critical Infrastructure Protectionhe explains that one of the most innovative features of the system is the use of a joint method based on mathematics known as strong feelings.

This method mimics the way people send messages to each other within their social networks to connect. When used on desktop computers, it allows AI agents to share what they’re seeing with each other and gradually adjust their observations to better understand what’s going on.