Welcome to Silicon Overdrive!
Artificial intelligence is changing everything, from how we write code to how we protect it. Yet, while AI continues to dominate conversations about productivity and innovation, one crucial topic rarely gets equal attention: security.
That was the focus of a recent webinar featuring Emil Munro, Senior Engineering Lead at Silicon Overdrive, who explored how AI is transforming cybersecurity, what the new threat landscape looks like, and how technologies like Microsoft Security Copilot are helping close the gap between defenders and attackers.
“Cyberattacks are’nt just faster today, they’re industrialized,” Emil began.
What was once the work of a lone hacker in a basement is now a global business model. On the dark web, attackers sell “as-a-service” offerings ranging from ransomware kits to denial-of-service attacks for as little as $60. In other words, anyone with the intent can hire a professional cybercriminal.
That means organizations can no longer afford to view cyber risk as a purely technical problem. It is a business issue that affects every department, from finance to HR.
“Security is everyone’s responsibility,” Emil emphasized. “It’s not just the IT team’s job anymore.”
For many organizations, multi-factor authentication (MFA) has been seen as the ultimate safeguard. But Emil warned that attackers have adapted and are finding new ways to bypass MFA through phishing and session hijacking.
Here is how it works: a user receives a convincing phishing email and clicks a malicious link. If that link opens in a browser where the user is already signed in, attackers can steal their session token — the small file that keeps a user logged in. From there, they can access systems without needing a password or MFA approval.
The solution? Conditional Access. Microsoft’s Conditional Access policies allow organizations to verify not just the user, but also the device, location, and risk level of each login attempt. If something does not look right, for example, a session token being used from a new country, the system can block access or demand re-authentication.
Ransomware has evolved from random spam to targeted, human-operated attacks. Emil shared how large-scale operations, sometimes entire call centers, are now devoted to scamming users, installing ransomware, and extorting organizations.
And because many companies lack proper backup or disaster recovery plans, paying the ransom often becomes their only option.
“It’s not a question of if you’ll face an attack,” Emil said. “It’s a question of when. That’s why preparation matters as much as prevention.”
His advice: adopt a Zero Trust mindset and continuously improve your defense posture. Never assume that one control is enough. Security is not static; it is a cycle of improvement.
Zero Trust is often misunderstood as a sign of distrust in employees. Emil offered a different view.
“It’s not about not trusting your people, it’s about not blindly trusting what’s trying to access your systems,” he explained. “A user’s laptop might already be compromised, and they would not even know it.”
Zero Trust operates on three core principles:
When implemented well, Zero Trust does not slow people down; it protects productivity by reducing the chance of a full-scale compromise.
While AI poses new risks, it is also creating powerful new defenses. Emil highlighted Microsoft Security Copilot as an example, a generative AI assistant built to support security teams.
The tool integrates with Microsoft Defender, Entra ID, Sentinel, and Purview to help analysts investigate incidents more quickly and effectively. It can:
For example, Emil demonstrated how simply asking, “Show me users who signed in from new locations in the past seven days” produces an instant query and report, saving hours of manual investigation.
But he was quick to add: “Security Copilot does not replace people. It assists them. It helps close the skills gap and makes security knowledge more accessible.”
As more employees use generative AI tools like ChatGPT, new risks are emerging. Emil cautioned against uploading sensitive files or proprietary data to public AI platforms.
“Even if the platform says it won’t share your data, it is still outside your organization’s control,” he said. “If that account gets compromised, so does your information.”
Instead, he recommended using enterprise-grade AI solutions, such as Microsoft Copilot for Microsoft 365, which operate within secure cloud environments and adhere to your company’s compliance rules.
Emil closed the session with a simple but powerful reminder: security is not a one-time project.
Attackers innovate constantly, and so must we. Implementing MFA, Zero Trust, and AI-powered tools is just the beginning. Ongoing training, monitoring, and adaptation are what truly keep organizations secure.
“Cybersecurity isn’t about fear,” Emil concluded. “It’s about awareness, preparation, and resilience. The more we understand the risks, and the tools we have, the better protected we all become.”
Silicon Overdrive is a Microsoft Solutions Partner specializing in cloud architecture, security, and digital transformation. Our team of engineers and architects helps organizations across Africa and beyond secure, scale, and modernize their IT environments using best-in-class Microsoft technologies.
Cybersecurity teams today are under constant pressure. By combining generative AI with Microsoft’s vast security intelligence, Security Copilot helps teams analyze threats, automate tasks, and make faster, more informed decisions.
The human factor in cybersecurity refers to the role employee behavior plays in either enabling or preventing security incidents. While human behavior is frequently the entry point for cyberattacks, research shows it can also become the strongest line of defense when supported by structured awareness training and modern security tools.
Unlike traditional security approaches that assume users or devices within a corporate network are safe, Zero Trust treats every access request as potentially risky, regardless of its origin. Every user, device, application, and connection must continuously prove its legitimacy before access is granted.
We use cookies to track visitors, measure ads, ad campaign effectiveness and analyze site traffic. We may also share information about your use of our site with 3rd parties. For more info, see, our Cookies Policy, our Privacy Notice. By clicking “Accept All” you agree to the storing of all cookies on your device. In case you don’t choose one of these options and use our website, we will treat it as if you have accepted all cookies.
We use cookies to track visitors, measure ads, ad campaign effectiveness and analyze site traffic. We may also share information about your use of our site with 3rd parties. For more info, see, our Cookies Policy, our Privacy Notice. By clicking “Accept All” you agree to the storing of all cookies on your device. In case you don’t choose one of these options and use our website, we will treat it as if you have accepted all cookies.