A growing number of organizations across the defense and security sectors are exploring the potential of artificial intelligence (AI) to enhance intelligence, surveillance, and reconnaissance (ISR) capabilities.
This rapidly evolving technology is transforming current intelligence and analysis operations by improving efficiency, reducing costs, and optimizing data management. Today, the role of AI’s deployment at the edge—on drones and sensors—is being explored to determine the impact it can have on reducing the burden and improving the performance of human operators responsible for harvesting insights from ISR platforms.
We sat down with Mark Rushton, Global Defense and Security Lead at VITEC, to discuss how AI at the edge is reshaping ISR platforms, the operational benefits it delivers, and the industry’s readiness to adopt these emerging technologies. He also highlighted VITEC’s role in driving AI advancements in the field.
QHow is AI currently influencing ISR platforms?
Mark Rushton: AI is already playing a significant role, particularly in command and control centers, where it helps manage IP video streams and make decisions about which footage to transmit. According to Forbes, around 77% of the market is exploring AI integration. The next major step is moving AI to the edge—on devices like drones and sensors—where it optimizes data transmission, reduces costs, and lightens the workload for human operators.
QWhat does AI at the edge mean, and how does it benefit ISR operations?
Mark Rushton: AI at the edge refers to embedding AI technology directly into data-collecting devices, such as drones or sensors, allowing real-time data analysis and decision-making before footage is transmitted. This saves bandwidth and cuts costs. For example, if AI detects a significant event during an ocean scan, it can alert the operator immediately, avoiding the need for hours of footage review.
Q How does AI at the edge improve ISR platform performance?
Mark Rushton: One key advantage of AI at the edge is that it doesn’t require additional hardware, which would increase the payload. By integrating AI into existing sensors and encoders, bandwidth usage is reduced, and the system can decide whether to stream in high or standard definition based on the importance of the footage. This reduces both operational costs and strain on infrastructure.
AI also supports efficiency, but critical decisions still remain in human hands. AI in ISR is designed to reduce the amount of data that operators need to process by highlighting what’s most important. While AI can autonomously decide whether to transmit footage, critical mission decisions will always be left to human operators.
Q Is the ISR community ready to adopt AI at the edge?
Mark Rushton: There’s growing interest — 27% of inquiries at a recent defense trade show were related to AI at the edge. Although AI adoption is still in the conceptual stage for many, manufacturers are actively working on integrating AI into sensors and encoders. The key challenge is ensuring the algorithms are reliable and capable of real-time decision-making.
The main challenge is ensuring that AI algorithms are robust enough to make accurate real-time decisions. Sensors and encoders need to work together seamlessly to flag and transmit important data. The technology will improve as AI systems are exposed to more data, but we are still in the early stages of development.
That said, some players in the market are already integrating AI, particularly in sensors and encoders that trigger data transmission. Full adoption is still evolving, but manufacturers are making progress. We expect AI at the edge to become more mainstream by the end of the decade.
Q How quickly do you see AI at the edge becoming mainstream in ISR?
Mark Rushton: It’s difficult to predict, as it depends on how quickly AI algorithms are developed and exposed to relevant data. Early-stage solutions could be operational within a year, but full saturation in the market will likely take until the end of the decade.
My recommendation is for leaders in the ISR community to focus on solutions that will grow as their needs evolve. AI won’t replace human decision-making anytime soon, but it can significantly improve efficiency. Organizations should ask the right questions now—such as about APIs, sensor compatibility, and AI triggers—so they are prepared for future developments. Look for solutions that are scalable and reliable.
Q How is VITEC supporting this trend toward AI at the edge?
Mark Rushton: VITEC has been thinking ahead for years. We developed region-of-interest coding six or seven years ago, which allowed high-resolution transmission in the center of the screen and lower resolution at the edges. Today, we’re focused on AI-enabled chips and improving APIs to ensure future-proof solutions. Our goal is to help organizations implement AI responsibly and cost-effectively.
