Artificial Intelligence in Space

Cybersecurity program from DARPA seeks to use AI to counter threats – LISA DAIGLE

This weeks article talks about the DARPA launched program called CASTLE (Cyber Agents for Security Testing and Learning Environments). Their intention is to use AI and machine learning (“reinforced learning”) to “automate the process of reducing vulnerabilities within a network”. All this software developed would be open-source for others to use. Per the article, “Tejas Patel, CASTLE program manager in DARPA’s Information Innovation Office, said of the program: “Attackers often have a better understanding of network vulnerabilities than defenders but it doesn’t have to be that way. Reinforcement learning may enable the creation and training of cyber agents that are much more effective than current manual approaches for addressing APTs in networks.” ”

While this is great for all networks it does open the possibility of application of this software onto a space system. Imagine a flight software that learns how to recognize that its being attacked. Or a command and control network with this implemented. Its hard to differentiate from a cyberattack or a vehicle anomaly for many satellite operators. Having a software to help could change the game.