The future of cyber security and AI in 2025

As we step into 2025, the landscape of cyber security and artificial intelligence (AI) continues to evolve at a rapid pace. This will have an impact on how you conduct business and protect data - both in a positive and negative way. 

Here are some key trends and predictions our experts see shaping the future and how they might affect you.

1. AI-driven cyber security solutions

AI is becoming a cornerstone in the fight against cyber threats. Advanced machine learning algorithms are being deployed to detect and respond to threats in real time, significantly reducing the window of vulnerability.  

This is great for your organisation as fewer attacks reduce the chance of you experiencing costly implications – fines, lawsuits, and brand reputation. For instance, Sophos’ research found that the average cost of recovering from a ransomware attack is $2.73 million. It also costs you a lot of time and hassle to resolve attacks. We know your time is precious, so this is a great benefit. 

For those of you considering creating AI-based software, The Hacker News highlights that NIST Artificial Intelligence Risk Management Framework and Google’s Secure AI Frameworks are just some of the frameworks you can use to avoid AI attacks and risks.

2. Increased sophistication of cyber attacks

The State of AI and Security Survey Report found that 25% of respondents expressed concerns that AI could be more advantageous to malicious parties. Perhaps they’re right to be so concerned, as All About AI’s report found that 51% of IT pros attributed successful cyberattacks to AI in 2024. 

Cybercriminals are leveraging AI to launch more sophisticated and targeted attacks. This includes the use of AI to automate phishing attacks and develop malware that can adapt to evade detection by your systems. This can leave your business very vulnerable. 

Deepfakes are also being used to trick organisations into sending money to fraudsters. Anyone can be targeted. From individuals being tricked out of £17,000 by a deepfake scammer as reported on Good Morning Britain, to large corporations like Arup in Hong Kong who lost $25 million due to a CFO deepfake video.  

Attackers can target AI platforms too. NIST has found that AI systems can malfunction when exposed to untrustworthy data, and attackers are exploiting this issue. 

Head to the bottom of this article for tips on how to protect your business against AI attacks and for tips on how you can detect a deepfake video.

3. Rise of autonomous security systems

Autonomous systems that can operate without human intervention are on the rise. These systems can identify and mitigate threats faster than traditional methods, providing you with a robust defence against cyber-attacks.  

Autonomous security will save you time, and maybe even money as you can consolidate resources. Plus, you’ll get peace of mind that your systems are always secure.  

But that does mean some people are concerned that AI will put them out of jobs. Goldman Sach predicts that 300 million jobs will be taken or degraded by AI.  

But as you’ll see on point five, humans - you - are still integral to the success of a business through working together with AI for enhanced efficiency. Lots of security vendors, such as ESET, Sophos and Acronis offer Managed Detection and Response (MDR) services that utilise the power of both AI technology and human expertise. These services provide you with extra resilience and support. If you’re interested in exploring how MDR protects your business, get in touch.

4. Focus on data privacy and protection

Data protection is no doubt front of mind for your business. And if it isn’t, it should be. With the increasing amount of data being generated, you need a heightened focus on data privacy and protection. New regulations and technologies are being developed to ensure that your personal and sensitive information is safeguarded.  

For example, the ICO offers guidance on AI and data protection. You can find information on ensuring transparency, lawfulness and fairness when using AI. 

Here’s an excerpt from the guidance: “If you use an AI system to infer data about people, you need to ensure that the system is sufficiently statistically accurate and avoids discrimination. This is in addition to considering the impact of individuals’ reasonable expectations for this processing to be fair.” 

This guidance and your own policies that you put into place, like making sure your own and your clients’ sensitive data is not used in public AI platforms will help you stay compliant and secure.

5. Collaboration between humans and AI

A survey by the Cloud Security Alliance found that 63% of the respondents agreed that AI would improve security in their organisation. And PwS research found that AI can potentially increase employee productivity by 40% by 2035. 

The future of cyber security will see closer collaboration between humans and AI systems. While AI can handle routine tasks and analyse vast amounts of data, your human intuition and expertise are still crucial for strategic decisions. Complex and ethical decision-making, relationship-building, empathy, and emergency problem-solving are some of the scenarios where your human skills are still required and are more effective than AI.

6. Emergence of Quantum Computing

Quantum computing poses both opportunities and challenges for cyber security. While it can potentially solve complex problems faster than classical computers, it also threatens to break current encryption methods, such as RSA and ECC, which secure much of today's internet traffic. This could compromise your sensitive data and communications according to Deloitte. Hence why many companies in that field are investing heavily in quantum-safe security solutions as noted in Global Risk Institute’s Quantum Threat Timeline Report. If you’re one of those businesses, talk to us to see if we can source the security solutions you need for that extra peace of mind. 

The shortage of people currently working in the quantum computing field challenges the adoption of quantum computing. Quantum Insider’s report highlights that the demand for quantum computing professions is expected to reach 10,000 by 2025, but the available workforce is half that. And 45% of organisations see a lack of skilled talent as a barrier to adopting quantum computing. For those of you wanting to utilise quantum computing more effectively for security, this might pose challenges such as delays getting the required resources and support.

7. Proactive threat hunting

Organisations are shifting from a reactive to a proactive approach to cyber security. This involves you actively hunting for threats before they can cause harm, using AI to predict and prevent potential attacks. Again, much like we mentioned on points one and five – this can save you time and costs. 

Lots of security vendors are integrating proactive AI tools into their solutions. For instance, ESET has an AI chatbot - AI Advisor - that helps your IT administrators and security analysts with interactive risk identification, analysis, and response capabilities.  

Want to explore AI Advisor or other AI-assisted security tools? We can help, just fill out the contact form below and our cyber security solutions specialist can help you. 

In conclusion 

As you can see AI can help you improve security and productivity. However, there are still many challenges that you need to manage carefully to mitigate the risks. All About AI found that 62% of businesses have adopted or are exploring AI for cyber security suggesting many see AI as a transformative tool for security management. 

For more information about the future of AI and cyber security, check out this article on ESET's blog.  

We also delve into this topic with ESET in this webinar recording:

How to protect your business against AI attacks 

You’ll need even more cyber security protection in place at every layer to protect your business against AI attacks. This will likely come with additional costs and time spent finding the right tools. But don’t worry, we’re on hand to help you find the best, affordable cyber security tools.  

And when it comes to using GPTs, The Hacker News suggests you can avoid risks by: 

• Not uploading sensitive data 

• Using instruction-based protection 

• Using OpenAI protection 

TechTarget also has tips on how to detect deepfakes manually and using AI.  

In the meantime, fill out the contact form below to book a call with our cyber security expert to discuss the security tools you need for multi-layered protection against cyber-attacks.