Rafael Núñez from Computer Weekly: Mythos, the new cyber intelligence that accelerates the security cycle
There are moments in cybersecurity where it's not a tool that emerges, but a change in scale that occurs
In the prestigious specialized publication Computer Weekly, Venezuelan expert Rafael Núñez Aponte, director of MásQue Seguridad, shares his insightful and high-level opinions on Mythos, the new cyber intelligence that accelerates the IT security cycle.
“There are moments in cybersecurity where it’s not a tool that appears (or emerges), but a change in scale that occurs. Claude Mythos Preview, presented by Anthropic, seems to be one of those moments.”
-It’s not a public product. It’s a contained experiment. According to its official website, it’s a model in the Preview phase, restricted to a closed group of organizations under the Glasswing program. There’s no open release date, and that alone speaks volumes.
“What they describe is technically disruptive. During testing, the model was able to identify previously unknown vulnerabilities in widely used operating systems and browsers. We’re not talking about trivial flaws, and some had gone undetected for years.”
-But the truly relevant aspect isn’t just finding vulnerabilities; it’s the ability to assist in building working exploits. In internal benchmarks, tasks that previously had marginal success rates began to be executed consistently. In some scenarios, even users without advanced security training achieved significant results by interacting with the model.
What they describe is technically disruptive
“For a penetration tester, or in the wrong hands, this means one thing: a radical reduction in attack time. Reconnaissance, correlation, and exploitation — phases that traditionally took days or weeks — can now be iterated in much shorter cycles. Not because the AI attacks on its own, but because it reduces the distance between vulnerability and viable exploitation. That’s the real game-changer.”
But this is where the public narrative falls short. Mythos isn’t just a threat. It’s potentially one of the most advanced defensive tools we’ve ever seen. The same model that can identify exploit paths can also anticipate them, simulate them, and allow them to be corrected before they materialize. That’s precisely the logic behind Glasswing: hardening critical systems before other actors develop equivalent capabilities.
“Cybersecurity has always been asymmetrical. The attacker needs to find a crack, while the defender must cover them all. With Mythos, that asymmetry begins to shrink because now both sides can operate with unprecedented speed.”
“The name isn’t accidental. ‘Mythos,’ in Greek, is the narrative that shapes reality. Not necessarily the truth, but the structure that makes it comprehensible.
Today, in cybersecurity, we’re entering a stage where machines don’t understand like we do, but they generate models of the world coherent enough to act upon it.” And when that happens, the risk is no longer in the tool itself: it lies in who learns to use it first and in whose hands it ultimately ends up,” concludes this article by the specialist.
Brief author bio: Rafael Núñez Aponte is a Venezuelan entrepreneur specializing in cybersecurity, digital reputation, and cyber intelligence. Internationally recognized for his work in ethical hacking and content strategies, he is known for promoting the democratization of digital tools and cybersecurity.
Taken from Computer Weekly / Written by Rafael Núñez Aponte
Reference image source: Steve A Johnson on Unsplash
Follow our news on Google! For current, interesting, and accurate information, click here to see all the content on Bitfinance.news. You can also find us on X/Twitter and Instagram
