The results of a survey show that, despite the threat situation and their concerns, the majority consider the AI to be of good quality and at least want to use it for personal use.
The rapid development of AI can currently be observed with your own eyes. The new GPT-2023 version of OpenAI was only released in March 4, and its opportunities and, above all, its risks are the subject of lively public debate. In a survey conducted by KnowBe4 in March 2023, 270 people were asked about their experiences and assessments of ChatGPT. A very interesting observation here is that despite the media hustle and bustle, “only” 37 percent of the participants have heard or read about ChatGPT in the news and only 18 percent have already used the AI themselves.
ChatGPT Threat Level
Around 46 percent consider the technology to be a very big or a certain threat. The respondents suspected a danger from ChatGPT, especially in the work in the company (23 percent) and in the field of politics and law (one fifth). The biggest concerns about using this technology can be found in the potential increase in the quality and quantity of cyber threats and attacks (31 percent) and the fear of losing one's job (29 percent).
However, this skepticism is mostly not reflected in the assessment of the quality of the chatbot's answers: More than three-quarters of the participants (77 percent) consider them to be very good to mediocre. In fact, both variables (assessment of the threat; evaluation of the quality) correlate with each other, an explanation for this average correlation could be the following: If one finds the performance of ChatGPT to be (rather) good, one is more likely to fear that this will be exploited by hackers or the technology could cost you your job. In contrast, a poor quality rating would mean that the AI is seen as less of a threat.
Beware of cyber attacks
This concern about an increase in cyber attacks is not unfounded: the security experts at Check Point were able to use ChatGPT to create high-quality phishing emails and malicious code for attacks to demonstrate the threat potential. The now constant availability of AI tools means that everyone, regardless of their knowledge, can easily access functioning attack techniques, and thus the threat landscape for the cyber world is expanding enormously and becoming more and more unpredictable. Overall, just under 40 percent of those surveyed would like to use the program for personal use. Even if fewer than eight percent assess the personal and family environment as a threatened area, many people do not seem to be completely convinced of the actual benefits.
Conclusion
Possible breaches of the GDPR by ChatGPT are currently being reviewed by the data protection authorities. So far there have been no formal complaints - unlike in Italy, where ChatGPT has already been blocked. For regulation, not only data protection supervision is required, but also politics. The majority of participants (42 percent) in the survey voted for the developers of the AI, 32 percent see politics and legislators as the responsible authority.
More at KnowBe4.com
About KnowBe4
KnowBe4, provider of the world's largest platform for security awareness training and simulated phishing, is used by more than 52.000 companies around the world. Founded by Stu Sjouwerman, IT and data security specialist, KnowBe4 helps companies address the human side of security by raising awareness of malware such as ransomware trojans, CEO Fraud and other social engineering scams sharpened. Kevin Mitnick, an internationally recognized cyber security specialist and Chief Hacking Officer of KnowBe4, helped design the KnowBe4 training and brought knowledge of his well-documented social engineering tactics. Tens of thousands of companies trust KnowBe4 to mobilize their end users as a last line of defense or "human firewall" in their quest to ensure their IT security.