Few months ago, my old friend, Gyongyi Sagi who runs Digitrendi a news portal in Hungary, asked my expert opinion about the risks of AI in cyber security, based on a study by one the vendors in this field.

To be in style, I have asked Chat GPT to summarize and translate this article, with the preliminary approval of Gyongyi, of course. Enjoy:)

This is an article that examines the connections between artificial intelligence (AI), especially language-generating tools like ChatGPT, and IT security. According to the article, AI-based threats are becoming more sophisticated and harder to detect, which could radically change security expectations and related education.

The article highlights that the cyber defence training programs of companies and institutions could be weakened by the emergence of AI-based solutions. One of the last effective elements of the training programs instructed users to pay attention to grammatical and spelling errors or texts that do not “sound good”. However, ChatGPT and other AI applications are capable of deceiving nearly 100% of people.

According to the experts in the article, we need to prepare for the increasingly dangerous waters of the internet, and we will need machines that recognize when other machines are trying to deceive us. Artificial intelligence has hammered the last nail into the coffin of end-user security awareness. The task is to teach users to be suspicious and to check communications that want access to information or inquire about our finances.

The Hungarian IT security expert featured in the article, József Gacsal, believes that the hype around ChatGPT strongly reminds him of the “fake video” hype a few years ago. He thinks this is just the beginning, not the end. According to him, the bad news is not that this is happening, but how media platforms handle these crimes.

According to expert József Gacsal featured in the article, generating fake human faces is not a new technology and has been used by certain spy agencies for years. Due to its simple use and free availability, it can be well used for creating fake profiles and fake companies.

The expert points out the dangers of fake recruitment campaigns running on LinkedIn and other platforms, which specifically target cyber security experts. These campaigns create databases about experts working in the field and can potentially provide sensitive information to foreign organizations.

However, József Gacsal believes that we should not despair and think that there is no point in security training. He believes that the paradox of cyber defence is still valid: we spend most of our expenses on tools while most problems are related to people. Until companies balance these factors, a radical improvement in information security cannot be expected.

According to the expert, the “bad guys” can continue to operate successfully, and we can see how much the world and individual national economies lose in favour of criminals. Therefore, it is important that we continuously develop and apply information security measures.

If you want to check out the original article by Gyongyi, click here:

A ChatGPT miatt sutba dobhatjuk, amit az IT-biztonságról eddig tanultunk? – digitrendi.hu