Real vs AI – The Face Off
Deepfakes have become a powerful tool attackers can deploy at every stage of the cyber kill chain.
From reconnaissance to execution, modern adversaries can now generate convincing identities, clone leaders’ voices, imitate employees on video calls with precision. Using open-source tools and AI models available on platforms like Hugging Face and GitHub, creating weaponised deepfakes is accessible to anyone with basic skills.
Traditional defences aren’t designed for an era where attackers can become someone inside your organisation.
Are you prepared to detect and disrupt a deepfake-enabled attack before it progresses?
In this live session, our cybersecurity experts break down how deepfakes are used to accelerate intrusions, bypass controls and amplify impact throughout the kill chain. You’ll see how attackers stitch together synthetic identities, how detection tools perform under pressure, and what practical steps organisations must take to harden their people, processes and technology.
Don’t just defend against social engineering.
Learn how to stop a full-chain deepfake attack.
What you’ll learn:
-
How deepfakes are deployed at stages of the cyber kill chain from reconnaissance and weaponisation to delivery, exploitation and exfiltration
-
Realistic attacker scenarios, showing how synthetic audio/video improves success rates
-
Validation techniques your team can apply instantly, even without specialist tooling
-
Incident response steps for deepfake-enabled attacks, including containment, escalation and post-attack communication
Why this matters — right now
Scope: Scattered Spider impersonated a Sys Admin employee using sophisticated voice authentication to execute access, lateral movement & a severe ransomware-based attack.
Impact: Impact: M&S have publicly stated the impact will be in the region of £300M, with weekly losses of £40M confirmed during the outage.
When one impersonated voice can cost hundreds of millions, recognising what’s real and what isn’t has never been more important.

