Anthropic’s new report is a wake-up call: hackers turned Claude into a near-autonomous espionage agent that handled 90% of the attack chain by itself. The future is officially here… and it’s terrifying.

We go deep on how they did it, why current defenses are cooked, and what defenders need to do yesterday.

Timestamps

00:00 – The scariest line Anthropic has ever published

01:17 – “Set it and forget it” – the new AI attack paradigm

04:04 – Breaking attacks into tiny, undetectable chunks

13:48 – Attackers flipped the script: 90% AI, 10% human

17:26 – Why defense has to 10x its speed right now

27:11 – SOC automation, log nightmares & the data problem nobody’s solved

33:18 – Thousands of API calls/sec + AI writing its own evasion logic

40:31 – Regulation debate: should frontier models be locked down?

51:38 – Back to basics… but the basics just changed forever

55:21 – Raw reactions: “Is this even real?”

58:09 – The silver lining (yes, there is one)

01:03:13 – When’s the next one coming?

Like if this freaked you out, comment your biggest fear for 2026, and smash subscribe — the AI cyber war just started.

#AICyberAttack #ClaudeAI #Cybersecurity #AgenticAI # infosec

http://cyberpodcast.net

Spotify: http://spotify.cyberpodcast.net

Apple: http://apple.cyberpodcast.net

X: https://x.com/dtfcyberpodcast

IG: https://www.instagram.com/dtfcyberpodcast/

Linkedin:

DTF: https://www.linkedin.com/company/dtf-cyber-podcast/

Damian: https://www.linkedin.com/in/damianchung/

Troy: https://www.linkedin.com/in/kosovotroy/

Fern: https://www.linkedin.com/in/fernrojasaz/

Business Inquiries: dtf at cyberpodcast dot net