Somebody's Watching Ep. 3 (If its free, You are the product)

Loading player...
1. The Mythos Dilemma – When AI Safety Fails
Anthropic’s powerful new model, Mythos, was designed to hunt vulnerabilities. But a breach through a third-party vendor. Raises the question what happens if one of the strongest AI models fall into the wrong hands? Federal officials and security experts warn that in the wrong hands, Mythos could cripple banks, hospitals, and governments.
2. Hypervisor Escape – The Crack in the Cloud
Imagine a hacker breaking out of one virtual machine to hijack an entire cloud server. That’s a hypervisor escape. This is one of the most devastating attacks in modern infrastructure. With threat actors hunting for these flaws, your “isolated” data may not be as safe as you think.
3. Free Apps & AI – The Hidden Cost
If you’re not paying for the product, you are the product. Free AI tools and free applications can collect, share, and sell your prompts including trade secrets or client data. From data monetisation, the price of “free” is higher than most realise.
13 May 9AM English Namibia Technology · News

Other recent episodes

Somebody’s Watching: Ep.2 Breaches that Teaches

Join Marcel and Michael as they break down what happened with the Standard Bank & Liberty Data Breach (2026) In early 2026, a threat actor operating under the name "Root Boy" conducted a three-week intrusion into the internal servers of Standard Bank and Liberty (its insurance and investment arm). The…
30 Apr 3AM 28 min

Obsidian Cyber Security Podcast

Join Marcel and Michael as they take a deep dive into Episode 1: Passwords That Open Every Door. Your password from that forgotten 2018 forum account just unlocked your bank,email, and crypto wallet. Password reuse creates a single point of failure that cascades across your entire digital life. Listen to…
29 Apr 4AM 12 min