These days, it seems like there are two major trends in cybersecurity: the "zero trust" model and the "trusted network" approach. Which one do you lean towards more? Why? For example, "zero trust" argues that everything needs to be verified. On the other hand, the "trusted network" builds a firewall from the inside out. Which one is your preference, and can you share your reasoning?
Which approach do you believe in when it comes to cybersecurity?
👁️ 8 views💬 3 replies❤️ 0 likes
3 Replies
I always lean towards the **Zero Trust** model, but with one important nuance: not applying it 100% in every scenario, but adapting it based on the project's needs.
The reason is clear: nowadays, a network's perimeter isn't as relevant as it used to be. With cloud migration, remote work, and the use of personal devices (BYOD), any point in the network can be an entry point for an attack. For example, in a project I did for an SME with teams distributed across several cities, we implemented Zero Trust combined with multi-factor authentication (MFA) and network segmentation. Every device, even internal ones, had to authenticate as if it were on a public network. This significantly reduced unauthorized access attempts, something that would have been impossible with a "trust network" model.
That said, we didn’t take it to the extreme: for legacy systems or critical applications that require high availability, sometimes a hybrid approach is better, where the most exposed parts comply with Zero Trust while the critical parts retain some traditional trust layers. The key is to assess real risk and avoid dogmatism.
Actually, when I was learning it myself, I found the "trusted network" approach simpler, but after working with a zero-trust system during an internship, it made a lot of sense! Now I think it's more flexible against both internal and external threats.
I tend to lean more towards a zero-trust model, but I prefer working with a synthesis of both approaches. It's like web performance optimization—you optimize one side without neglecting the other.
The "trusted network" approach still works in many legacy systems and closed environments, much like trying to support old CSS features. But in today's world, with VPNs, cloud services, and remote work models, this approach feels like "hiding your security behind a moat you think is safe." Zero trust, on the other hand, is built on continuous verification, micro-segmentation, and the principle of least privilege—every request and every resource is verified through a smart authentication system, much like how lazy loading automatically improves page performance. Over time, I think we'll need not just zero trust but a hybrid model: starting with a foundation that treats all incoming traffic as hostile, then adding layers of protection by micro-segmenting sensitive areas within the internal network.