Is cybersecurity the sixth sense of AI?

Publish date:

As humans, we tend to “feel” when something is not quite right. This is our instinct, or sixth sense.

There’s evidence of a sixth sense developing in Intelligent Automation that could help us reframe our approaches to some of today’s most serious security threats.

In my last article, I introduced The five senses of Artificial Intelligence (watch, remember, think, talk, act) as a useful framework for businesses exploring automation solutions. My argument being that it isn’t helpful to think of AI as a single thing—instead, it is the integration of several different solutions and capabilities. You need to plan for all five senses working together if you want to replicate human intelligence.

When this happens, something special will also start to occur: machines will develop an instinctive response to external factors, with a human-like intuition that can’t be programmed by rules. This will create huge potential for businesses, particularly in complex areas like cybersecurity.


Automation has played a significant role in cybersecurity for many years. It’s done a lot of helpful heavy lifting in terms of monitoring events (watch), detecting threats (think) and escalating risks (talk). Primarily, this has been in an assistive role—filtering information to human security staff to take action.

Yet, as the volume and diversity of digital data has exploded, particularly via IoT sensors, security teams simply don’t have the capacity to manage all these threats. Speed of response has become critical in the fast-moving cyber-risk landscape. Which is why developments in automation have focused on enabling machines to do even more, using the knowledge of previous security events (remember) to make appropriate proactive responses to new threats (act).


By bringing together these five senses, machines are now starting to learn and proactively redesign themselves to resist emerging threats. For example, NASA is exploring cyber reasoning systems that identify threats and take protective action at “cyber speed,” beyond the ability of humans.

Research is also underway to look at “using artificial intelligence to analyze and protect urban infrastructure,” which could help future smart cities to identify hacking threats and build countermeasures automatically.

We’ll see more of this. With machines talking to other machines, working together to identify and act on threats, even learning from each other.


While some of these developments are still in their beginnings, there is evidence of the sixth sense in action today. For example, intelligent machines are now working alongside security professionals as their expert assistants—not just flagging threats, but advising on risk levels and available options.

They do this by consolidating relevant structured data related to the threat and presenting the findings to the human in rapid speed. But their advice also incorporates unstructured information, including blogs, forums, best practice papers, and images. The result is something that resembles well-informed, well-rounded suggestions from an experienced colleague, not just a factual escalation report.


The big change here is that insight is moving from the domain of humans to the domain of machines. They have a virtually limitless capacity to access and process data, and when this comes together with the other senses of Intelligent Automation, the result can feel almost uncanny. But as Arthur C. Clarke once said, “Any sufficiently advanced technology is indistinguishable from magic.” Intelligent Automation is inevitably becoming that.

To read more about Capgemini’s approach to cybersecurity, including new ways to control and secure your assets, visit our hub page.

In my next article, I’ll be considering if Intelligent Automation could help us rethink some of the fundamental tenets of business process management. Could we be seeing a turnaround for lean thinking?

Related Posts


Can AI save the security operations center?

Geert van der Linden
Date icon November 30, 2021

AI technologies can also perform threat modeling and impact analysis, activities that...


Mastering the critical art of cybersecurity in Automotive

Geert van der Linden
Date icon November 3, 2021

Steps OEMs need to take in order to get ahead of cyber threats


The difference engine

Date icon November 3, 2021

Capgemini’s Digital Global Enterprise Model platform reflects the unique and individual...