Cybersecurity
Post LinkedIn lead magnet · Cybersecurity
🤯 AI is BREAKING traditional cybersecurity in ways we aren't ready to fix. I love the show Stranger Things. This concept of "upside down" has got me thinking how AI has made cyber go “upside down”, in at least 3 ways since GenAI became a thing. Let me explain. [1] AI is BREAKING the traditional ways of identification and verification in today’s world. Deepfakes have not only become commoditised, but they are mostly still unregulated and they have blurred entirely the lines between fake and real, fuelling even further enterprise risks like disinformation, electioneering, fraud, etc. [2] AI is BREAKING the boundaries of risk management for businesses. AI is not just any tech. Even without AGI, AI has issues like ethics, biases, etc. Things can go wrong both at the data and the model level. In all my roles building and leading enterprise risk management, two key things my team and I always did... a) integrate threats into the risk management framework and b) integrate tech risk into enterprise risk. No siloed approach. Yet, I still see so many companies have zero idea of what impact their threat landscape has on their enterprise risk, especially when it comes to AI risks. [3] AI is BREAKING the concept of what once meant "insider threat" Agentic misalignment is real. Even if under certain circumstances, an agentic AI going rogue is highly plausible. You don't need to wait for AGI to happen for AI to go rogue. Your insider threat is no longer just a grumpy employee but now also your agentic AI managing your damn workflows that are critical to your business decisions. Add to that, most companies don't even have insider threat as a part of their threat analysis and risk management. While most basics are still valid, in this world of AI, we do need to re-think and re-asses a lot of those "broken" traditional cybersecurity measures. I published my cybersecurity model in 2020, where I predicted that the next waves of cybersecurity adoption will be adaptive and predictive in nature. In the world of AI, it definitely will need to be. As we are ending this year, here's my BOLD prediction for 2026: — We will see way more "predictive" security measures going forward, e.g. AI helping us hunt threats in real-time (independent of sophistication level and despite the non-sophisticated nature of many attacks). This is going to be massive, as most companies still don't understand their threats, and its impact on business risk. — What's your prediction / PoV for 2026 on how AI / cyber is moving? 🚨 Here's YOUR CHANCE. Drop your comments below with your 2026 prediction (YOUR PoV), follow Armis and I’ll pick three WINNERS that will get a cool merch for free. Are you game? UPDATE: This was a very difficult decision with so many insightful PoVs/predictions. The three winners are Jake, Oge and Dr Mike! A big thanks to everyone for being an epic part of my community. Partnered with Armis #cyberthings #artificialintelligence #cybersecurity #leadership
Mécanisme lead magnet
Drop your comments below with your 2026 prediction (YOUR PoV), follow Armis and I’ll pick three WINNERS that will get a cool merch for free.
