The pros and cons of AI in cyber security

The use of Artificial Intelligence (AI) is growing. Organizations are adopting the technology in different areas throughout.

AI has many benefits, but there are growing concerns about its usage including its role in the cyber security sector. A review of both pros and cons can provide guidance and raise awareness of the benefits and risks AI presents in cyber security.

The Pros of AI in Cyber Security
AI allows us to recognize data patterns that aren’t readily visible to a human. These anomalies may reveal efforts to steal or compromise organizational data. AI can also be used to help craft policies by simply prompting AI to reword technical training concepts and converting them into concepts that are more easily understood by the non-technical person.

By automating response activities, AI can also improve incident responses by identifying alerts and providing faster responses to suspicious or malicious activities.

AI can also be trained to take certain autonomous actions by shutting down or taking a machine offline to short-circuit an attack and do it much faster than traditional methods.

The Cons of AI in Cyber Security
Perhaps the most pervasive drawback of AI is that it provides the less skilled attacker, or even an attacker with no programming skills at all, with the ability to create or adapt malware.

AI requires organizations to diligently protect their confidential corporate information. AI tools themselves have become targets of bad actors and can be breached. A compromised AI can be made to provide false information or take wrongful actions. So, any information an organization inputs into an AI tool is at risk of exposure.

Deepfakes are also a risk. Content created by AI tools including images, text, videos, voice recordings, etc. are used to create very believable disinformation or social media attacks that compel the public to act without thinking.

AI can also be used for password guessing to optimize attacks. AI can gather information from the internet (i.e. social media) about a person or organization to create the most likely password options to use in their attacks. Since AI is good at analyzing data, it can also find security gaps that may be unrecognizable to humans and can be used to optimize attack vectors.

The Future of AI
We expect AI to improve, get smarter, and become more accessible over time. AI will free humans to be more creative and focus on higher-order tasks while the more basic tasks such as data entry, content generation, and customer support can be performed by AI tools.

There are ways to mitigate the risks of using AI. One way is to create your own private AI instance. Another way is to verify or validate output for false information to guard against infringement of intellectual property. Ultimately, we expect to see tools come into the marketplace to help organizations manage the use of AI and minimize the risks.

If you would like help navigating these new tools and how you can maximize them for your organization, please feel free to reach out to us.

Editor’s note: Daniel Weatherly is Director of Security Services for Mainstream Technologies, Inc. The opinions expressed are those of the author.