This week, AI became the Hacker
What happened between OpenAI and HuggingFace is a wakeup call for every technology leader
An AI model, during a cybersecurity evaluation, found a way out of its testing environment and reached another company’s infrastructure. Not because it was “evil.” Because it was relentlessly trying to achieve its objective.
That made me pause.
For years, we’ve been asking: “Will AI replace people?” Maybe we’ve been asking the wrong question.
The better question is: Are we building AI faster than we’re building the guardrails to govern it?
Technology doesn’t create risk
As someone who has spent years in cybersecurity, this incident doesn’t make me fear AI. It reminds me of something we’ve always known.
Technology doesn’t create risk. Uncontrolled capability does.
The next generation of cyber defense won’t be about having the smartest AI. It will be about having the strongest governance, the best visibility, and the discipline to keep powerful systems accountable.
One sentence stayed with me
The future of cybersecurity won’t be decided by who builds the most intelligent AI. It will be decided by who governs it the best.
If an autonomous AI agent made a critical security decision inside your organization tonight… Would you know?
I’d love to hear your perspective.
Instincts built hunting signals at sea, sharpened for two decades against digital adversaries. Not machine-written. Machine-proofread, comma by comma. Everything else, the noticing, the arguing, the getting it wrong sometimes, is mine. The bytes are for everyone; the opinions are only ever mine.
“Until the lion learns how to write, every story will glorify the hunter.” This is the lion's version.