The Case for and Against AI’s Role in Cybersecurity: Is AI a Hero or a Villain?

Love it or hate it, AI is here to stay – especially if you’re in the cybersecurity industry.

Yet AI’s role in that market (one already worth $31.1 billion in 2024 and set, thanks to AI’s rapid and ongoing evolution, to be worth almost five times that by 2030) is a controversial one.

Why? Because the forces of AI aren’t being harnessed solely for the protection of the world’s internet users, businesses, and data banks – but for their destruction, too.

Alright, that’s perhaps a little dramatic. But the fact is, 75% of security professionals surveyed in 2023 reported an increase in cyberattacks over the past year; of these, 85% blamed AI.

So, does AI represent a positive development in the internet’s eternal struggle against theft, fraud, and criminal deception? Or simply another tool for bad actors to exploit online?

Below, we explore each side of this hot debate. We’ll make the case both for and against AI’s role in cybersecurity, then explore the biggest cyber threats named by AI to look out for in 2024 – and what you can do to protect you and your organization from becoming a target.

The Case for AI in Cybersecurity: The ‘Good’

First up, we’re making the case that AI is a force for good in the cybersecurity space.

Read on for our top three reasons why.

1. AI Can Combat All Types of Cyber Crime

Over a third (34%) of organizations are already implementing or using AI cybersecurity tools to stave off online threats. So these tools must be doing something right…right?

Right. One study found that, when it comes to identifying malicious scripts, AI was able to detect 70% more bad code than traditional methods, with an accuracy rate that dwarfed that of conventional threat-detection techniques by as much as 300%.

Through techniques such as rapid incident analysis and advanced persistent threat detection (APT) – not to mention the ability to detect and neutralize advanced phishing threats – AI is able to combat cyber crime in a way, and at a scale, that solely human teams of experts can only dream of.

Speaking of which…

2. AI-Enabled Cybersecurity Saves Businesses Time

Empathy; creativity; emotional intelligence; moral and ethical judgment calls. There are a lot of fields in which humans still trump AI.

Work rate, however, isn’t one of them.

Aided by the adaptable capabilities of machine learning algorithms – which continually learn from inputted data to make decisions at ever-evolving rates of accuracy – AI-powered cybersecurity tools are able to get through a staggering amount of work.
Whether that’s a bank monitoring transactions for potential red flags or a website keeping tabs on suspicious spikes in traffic, AI is capable of workloads unthinkable by human teams. (It helps that AI doesn’t need to eat, sleep, go on vacations, or take toilet breaks, too.)

It’s unsurprising, then, that seven in ten (69%) enterprises believe AI is a necessary part of their cybersecurity setups – simply due to the growing number of threats they’re unable to get to. Another study, this time of Security Operation Centre staff, suggested that employees spent as much as a third of their time investigating and validating false positives in 2023.

What’s more, 80% of staff in this same sample also reported that the manual effort involved in these tasks scuppered the speed and effectiveness of their processes; and slowed their overall threat response times.

3. AI Facilitates Faster Detection of Data Breaches

Data breaches – incidents in which protected or confidential information is accessed by hackers after a cyber attack, or through a phishing scheme – are a huge threat in 2024.

In 2023, 2,814 data breaches saw more than 8.2 billion records exposed, with huge swathes of sensitive data – including names, addresses, phone numbers, social security numbers, credit card details, and even health records – pilfered by cybercriminals. Of course, this is damaging for a business not only from a brand image perspective, but a monetary one – data breaches cost organizations, on average, a jaw-dropping $4.45 million in 2023.

Fortunately, though – and despite the fact that AI, by enabling more sophisticated phishing attacks, is facilitating data breaches in 2024 – AI is also helping safeguard organizations from the financial and reputational consequences of data breaches.

For example, one survey from IBM demonstrated that, in 2023, data breaches that hit organizations with fully deployed AI-driven cybersecurity solutions cost them $1.8 million less, on average, than companies without a similar setup in place. IBM’s research also found that these less well-equipped organizations took an average of 100 days more, vis a vis companies with AI-powered defenses, to identify and contain data breaches.

To put a financial figure on this, the companies with AI-enabled security only cost organizations $2.65 million in 2023 on average. And if that still sounds like a lot, it’s over 40% less than the losses organizations without AI cybersecurity absorbed ($4.45 million in 2023).

Not a bad investment!

The Case Against AI in Cybersecurity: The ‘Bad’

Every good case deserves a good rebuttal – so what are the arguments against AI’s prosocial involvement in the cybersecurity space?

We’ve summarized our top three below.

1. AI’s Role in Cyberattacks of Increasingly Alarming Scale

By enabling more sophisticated and scalable cyber attacks – including those of the Distributed Denial of Service (DDoS), brute-force, zero-day, and cross-scripting varieties – AI is becoming an increasingly potent and prolific weapon for online criminals.

How? By making attacks quicker, smarter, more effective, and able to be implemented at scale, not to mention at relatively low cost.

As one UK government report put it, AI means “faster-paced, more effective and larger-scale cyber intrusion”. Looking at the size and ferocity of some recent attacks – a DDoS attack on Google, for example, which culminated in an unprecedented 398 million requests per second – and it’s hard to argue with the Home Office’s alarming assessment.

2. Concerns Around Ethics and Privacy in AI

Asked to name their most pressing concerns about AI implementation in a survey, cybersecurity experts’ top worry was around privacy, with 39% citing “increases in privacy concerns” as a blocker to AI adoption (Deep Instinct, 2023).

Of course, to anyone familiar with the insatiability of machine learning models – which, by their very nature, demand a near-constant input of information to do their jobs – these privacy concerns come as no great shock.

Where, after all, is this data coming from? Who’s responsible for its collection, storage, and use – and what principles are governing and regulating all this handling of personal information? With concerns around the data-privacy policies of tech giants – such as Google, Zoom, and Meta – continuing to simmer and fester, AI-related privacy fears are only set to grow in 2024.

3. AI Has Experts – of Both the Human and Artificial Kind – Worried

In a Deep Instinct-led survey of security professionals in 2023, nearly half (46%) of the respondents expressed a fear that generative AI (the type that’s able to create content, such as deepfakes or writing; ChatGPT is one example) will make companies more vulnerable to cyber attacks than they were in a pre-AI world.

Across the board, AI is clearly something that has human experts on edge, with 37% of cybersecurity experts citing ‘undetectable phishing attacks’ as their top concern; and 33% listing ‘the volume and velocity of cyber attacks’ as their chief worry (Deep Instinct, 2023).

Yet it’s not just human intelligence growing uneasy around AI’s role in facilitating cyber threats. It’s artificial intelligence, too!

Recent research by Techopedia asked five different AI language models – ChatGPT, Llama, Claude, Perplexity AI, and Bard – what the biggest threats on the internet were. Of these, four directly name-checked AI-powered cyber attacks as the online space’s key cause for concern, with ChatGPT stating: “Phishing techniques have become increasingly sophisticated, often leveraging AI and machine learning to create highly convincing fake messages and websites.”

On that note – what were the biggest AI-identified cyber threats to look out for in 2024?

Let’s take a look.

The Biggest Cyber Threats in 2024 – and How to Stop Them

Before we wrap things up, here’s a quickfire list of just a few of the top cyber threats to look out for in 2024 – as identified by the five AI language models Techopedia ‘consulted’ as part of its research – with a note on how you can arm yourself, or your organization, against them.

Threat: Sophisticated, AI-enabled phishing attacks

Solution: Education! Learn more about the latest phishing trends and statistics, and – if you run a business that employs staff – train your team around what phishing scams are and how to spot them. This could take the form of cybersecurity awareness courses or phishing simulations, which mimic a real phishing attack to test your employees’ knowledge.

Threat: AI-assisted brute-force attacks

Solution: Create strong, unique passwords. With AI increasing the speed and scale at which bad actors can perpetrate brute-force attacks, making your passwords impenetrable is vital. (Any password containing numbers only is instantly guessable with AI, and even complex ones containing eight characters can be cracked by AI in as little as – gulp – seven hours.)

Threat: Deepfakes and AI-generated misinformation (a.k.a. ‘fake news’)

Solution: If possible, verify everything you see online, Check the facts, double-check the sources, and be skeptical of anything that doesn’t quite look right.

Threat: Compromised data security and ‘snooping’

Solution: Use a VPN (Virtual Private Network) to anonymize your browsing. VPNs channel your web traffic through a secure, encrypted tunnel – often masking your IP address with one from your choice of locations around the world – to prevent unwanted ears eavesdropping on your online activity. Unfortunately, it’s not just cyber criminals that can, or will, do this: it’s your ISP (Internet Service Provider) too, and, most likely, your own government.

That’s it from us! To dive deeper into a huge array of topics – including how AI is boosting ransomware, the top scams to look out for in 2024, and our list of the most heavily censored countries in the world – stick around: you’ll find it all right here on Digital Information World.

Read next: Earning Customer Trust, One Review At A Time: The Us Companies With The Most 5-star Reviews On Trustpilot
Previous Post Next Post