A small business does not need to be on a hacker’s target list. A single security gap may be enough for automated programs to find their way into its systems.
Safe IT-Solutions in Vordingborg sees this among its customers. According to CEO Matias Aabye, AI can make attacks more effective because they can be carried out on a large scale without a person sitting behind every attempt.
- With AI, the algorithm can attack and search for security weaknesses without a human being directly involved. It is a much more efficient way of attacking, and AI does not care whether it hits a large or a small company, he says.
Safe IT-Solutions regularly sees automated attempts to break into customers’ systems, and the attacks are not necessarily the result of hackers deliberately targeting a particular company.
- These attacks are different precisely because they are not aimed at one specific company or another. They are basically bots attacking and constantly checking whether there are weak passwords or other vulnerabilities that can be exploited, says Matias Aabye.
Part of the solution
AI also features more prominently in PwC’s latest Cybercrime Survey. The share of Danish companies surveyed that see new technologies such as AI, cloud and IoT as one of the biggest cyber threats has risen from 26 to 52 per cent in a year.
- It is worth noting that Danish companies are investing heavily in AI at the same time as more of them see the technology as a growing cybersecurity risk, says Mads Nørgaard Madsen, partner and head of Cyber Excellence at PwC.
The survey measures companies’ perception of the threat, not the number of attacks.
At the same time, AI has moved from fourth to second place among the areas companies prioritise for investment in cyber and information security. It is now a priority for 51 per cent, up 25 percentage points from the year before.
- The solution is not to keep AI out, but to use the technology within clear security frameworks, with ongoing risk assessments and employees who are equipped to use AI responsibly and securely. Security needs to be part of the AI solution from the outset, says Mads Nørgaard Madsen.
