ÌÇÐÄÆƽâ°æ

According to a 2025 survey, 30 percent of respondents highlighted that data leaks were their greatest concern regarding the impact of generative artificial intelligence (GenAI) on cybersecurity. In addition, ** percent of respondents were concerned about the advancement of adversarial capabilities.

AI-powered malware...

With the launch of OpenAI’s ChatGPT in November 2022, concerns have been rising around its possible usage in cyber crime. Trained to create human-like texts in a shorter time without spelling errors, phishing e-mails written by ChatGPT would consequently be harder to detect, for instance. In addition, there is growing concern about AI-powered malicious software, commonly known as malware, as deep learning algorithms would allow hostile actors to target specific victims and remain undetected until specific conditions are met.

... Versus AI-powered cybersecurity

Risks aside, the advantages brought by AI to cyber criminals can also bolster cybersecurity. In particular, generative AI-powered solutions can search through vast amounts of data to identify abnormal behavior and detect malicious activity. Looking forward, companies will have to adapt and stay up to speed so that generative AI does not end providing overall cyber advantage to attackers.