Executive Interview: Separating fact and fiction: How AI is changing the cybersecurity landscape

Artificial intelligence (AI) is becoming a crucial tool in cybersecurity, helping organizations detect and respond to threats faster than ever before.

However, a recent survey by Sophos found that 89% of IT leaders believe flaws in generative AI could introduce new risks1. With cybercriminals also leveraging AI to refine their attacks, businesses must understand both the opportunities and challenges AI presents in cybersecurity.

This article will explore the balance that needs to be struck for long-term success, with insight from Sophos’ Principal Research Scientist Chester Wisniewski.

The Changing Cybersecurity Landscape

Cyberattacks are evolving at an alarming pace. The time between a breach and a damaging attack is shrinking. In the past, organizations had weeks to detect and respond to a cyber threat. Today, that window is much smaller.

Sophos’ Chester Wisniewski said: "The amount of time it takes to breach an organization, find damaging information that they may want to use for extortion or do encryption or all the different modes of operation the criminals are deploying, the window was shrinking, right?

“We used to have often a couple of weeks from the time of breach until something bad happened. And that time has been coming down and down and down. And now we're in kind of this... you know, worst case scenario of five to eight days."

So, why is this happening? One factor is that attackers are using automation and AI to optimize their tactics, making speed a critical factor in cybersecurity defense. In fact, a recent survey from Darktrace revealed that 74% of security practitioners acknowledge the significant impact of AI-driven cyber threats on their organizations, yet 60% fear they are not adequately prepared to defend against such attacks2.

To keep up, businesses must leverage AI-driven security tools that can process large volumes of data in real-time. While human analysts excel at detecting complex threats, they can’t match the speed at which AI can analyze logs, identify anomalies, and flag potential attacks.

How AI Strengthens Cybersecurity

AI is already embedded in many cybersecurity solutions. Machine learning algorithms power threat detection in endpoint security tools, helping businesses identify malware, phishing attempts, and suspicious network activity. AI also assists in automating repetitive tasks, such as triaging security alerts, scanning emails for phishing attempts, and analyzing malicious links.

One of AI’s biggest advantages is its ability to reduce the workload on security teams.

“Triaging phishing emails, analyzing logs—these are tedious tasks that lead to burnout,” said Wisniewski. “AI can handle the repetitive work, allowing analysts to focus on more complex threats.”

However, AI is not infallible. Generative AI tools, while powerful, are not always accurate and can be manipulated by attackers. That’s why human expertise remains essential. Security professionals must supervise AI-driven tools to validate findings and make final decisions.

Challenges in AI Adoption

Despite its benefits, AI adoption in cybersecurity comes with challenges. Developing AI-driven security solutions requires extensive expertise, large datasets, and significant investment. There are also external factors to consider, like compliance or sustainability.

While large cybersecurity firms like Sophos have been integrating AI into their products for years, many organizations lack the resources to build their own AI models.

Another issue is that many organizations simply find building their own AI models too complex, expensive or resource intensive. Companies like Sophos help that pain point by offering prebuilt AI-powered security solutions that businesses can implement without requiring in-house data scientists.

These tools provide advanced threat detection without the complexity of developing AI models from scratch and bring their own expertise and support to the table.

The Future of AI in Cybersecurity

Looking ahead, Wisniewski believes collaboration between private companies and law enforcement will play a key role in strengthening cybersecurity. Increased information sharing and coordinated efforts will help defenders stay ahead of cybercriminals.

“The last three to five years have seen major progress in taking down cybercriminal networks,” he said. “If we continue improving coordination and response times, we can put more obstacles in their way and better protect organizations.”

AI is transforming cybersecurity, but it works best when combined with human expertise. Businesses must implement AI strategically, ensuring that it enhances - not replaces - security professionals.

With the right approach, AI can speed up threat detection and response, making cyber defenses stronger, more efficient, and ultimately transforming security efforts for the better.

Want to hear more about how AI is changing the world of cybersecurity? Watch the full interview between Syndey Neely (CRN) and Chester Wisniewski (Sophos) here .