Binary Defense ARC Labs to Host ThreatTalk on AI’s Role in the Cybersecurity Arms Race

https://sg.finance.yahoo.com/news/binary-defense-arc-labs-host-110000417.html

Publish Date: 2026-06-15 07:00:00

CLEVELAND, June 15, 2026–(BUSINESS WIRE)–Binary Defense, the trusted Managed Detection and Response (MDR) and enterprise defense provider, today announced ThreatTalk Episode 11: From Threat Intel to the SOC: How AI Is Accelerating Both Sides of the Fight. This live webinar takes place Thursday, June 18, 2026 at 2:00 PM EDT and is open to cybersecurity professionals at organizations of all sizes.

Produced by ARC Labs, Binary Defense’s threat research division, ThreatTalk is a practitioner-led webinar series built for security teams who need more signal and less noise. Episode 11 tackles one of the most urgent questions in the industry: how AI is simultaneously enabling attackers to move faster and giving defenders the tools to match them.

What Attendees Will Learn

The threat landscape has changed dramatically in three years. Financially motivated actors who once used AI as a basic productivity aid are now operating purpose-built attack platforms purchased on dark web forums, including tools like WormGPT, FraudGPT, and SpamGPT that automate phishing, generate polymorphic malware, and enable deepfake-based fraud at scale.

The next phase is already here. Experimental malware like PROMPTFLUX actively queries AI APIs mid-execution to rewrite its own source code on an hourly basis, ensuring every variant evades static detection. PROMPTSPY, an Android-based malware family, uses an LLM-driven automation agent to navigate victim devices autonomously, interpret UI states, and block uninstall attempts with invisible overlays. And threat hunters are watching a new category emerge: malicious AI “skills” built for platforms like OpenClaw that extend attacker capabilities inside victim environments without detection.

Looking further out, the panelists will walk through three converging threats: social engineering via malicious llms.txt files designed to manipulate how AI agents perceive legitimate websites; prompt injection attacks targeting SIEM,…

Source