As language models (LLMs) in enterprise applications continue to grow, ensuring data privacy while maintaining response accuracy becomes crucial. One of the primary methods for protecting sensitive information is data masking. However, this process can lead to significant information loss, potentially rendering responses from LLMs less accurate. How can this loss be measured?

The “Russian Laundromat” scheme is a notorious example of how complex and far-reaching financial crimes can be. Running from 2010 to 2014, this scheme funneled between $20 billion and $80 billion out of Russia, using a network of shell companies and global banks, many of which were based in Moldova and Latvia. This was not a small operation — it involved around 500 people, many of whom were wealthy Russians, and relied on thousands of transactions to launder the money.

Top tips is a weekly column where we highlight what’s trending in the tech world today and list ways to explore these trends. This week, we’re going over five steps to building a robust cybersecurity mesh. The last few years have seen computing models and network infrastructure become increasingly fragmented and distributed.

Did it ever cross your mind to ask if your password can defend your sensitive info on the web all by itself? In the digital world of today, where cyber threats are rapidly developing, the traditional way of protecting the system with single password has become more and more insecure. Therefore, it is the user who is more exposed to such cyberattacks. Just to mention, more than 81% of data breaches are associated with weak or stolen passwords.

Email remains a critical communication channel for businesses of all sizes, but it also presents significant data security risks. Data loss prevention (DLP) for email is an essential component of any robust cybersecurity strategy that can help your organization safeguard sensitive information from accidental leaks, malicious insiders, and external threats.

As offensive security specialists for over 10 years, we have tested countless organisations who believe their SIEM, EDR or MDR provider offers them comprehensive defense, only to find them lacking in fundamental areas. From our experience, some “traditional” in-house, yet adequately resourced, Security Operations Centres (SOCs) can still provide a robust defense, while others struggle to stay on top of emerging threats.

As companies handle increasingly sensitive information, securing data in motion—the data actively transferred across networks or between devices—has become an important priority. Whether moving through corporate networks, across mobile devices, or stored in the cloud, this data is particularly vulnerable to threats from malicious actors, insider threats, and unauthorized users.