CTI-REALM is Microsoft’s open-source benchmark that evaluates AI agents on real-world detection engineering. It measures whether an agent can take cyber threat intelligence (CTI) and produce validated ...

RunSybil’s AI agent continuously tests live applications for vulnerabilities, aiming to replace traditional penetration testing and bug bounty programs.

AI could soon be able to hijack satellites and cause them to collide with other spacecraft, triggering a cascade that could render Earth orbit unsafe.

As artificial intelligence (AI) becomes woven into the fabric of daily life – powering automation, analytics, and decision-making – it simultaneously opens new vulnerabilities for attackers to exploit ...

Maria Sumnicht was a young software engineer at NASA Ames Research Center in 1988, when the Morris Worm — considered one of the first cyberattacks in the nation — infected about 10% of the internet, ...

Organizations increasingly are turning to LLM-based autonomous agents to detect and prevent threats to their networks. As cybersecurity threats grow, artificial intelligence (AI) is revealing itself ...

HOUSTON, TX, UNITED STATES, March 17, 2026 /EINPresswire.com/ — Tantalum Security, a bleeding-edge adversary simulation and active-defense cybersecurity firm, today announced the official launch of ...

Will AI make cybersecurity obsolete or is Silicon Valley confabulating again?

In 2026, CEOs at major cybersecurity vendors are continuing to accelerate their deployment of AI across their product portfolios, with a strong focus on bringing agentic-powered security to the fore.

Quantro Security, Inc., an AI-native company redefining cyber defense through autonomous agentic AI capabilities, today emerged from stealth to unveil its first product – VM.Analyst. This AI Agent is ...

Even with the strongest technical defenses, humans remain a critical factor in cyber security.

Adaptive Security, the leading provider of AI-powered social engineering prevention solutions, today launched free public ...