One of the hottest news in the AI field is the launch of GPT-4 by DeepAI. As the developers state, outperforming the previous version is not the only improvement made. Now, GPT-4 is able to produce much safer and more factual replies, preventing its misuse.
Still, it doesn’t mean that it can not be used for destructive purposes, starting from academic misconduct and up to producing malicious code for hacking purposes. The simplest, lying on-the-surface way to deal with this problem is to use GPT4. The person can detect the AI-generated text (like ChatGPT) by asking about the probability of this from AI, and similarly, cybersecurity can be improved with the use of GPT-4.
Credits: Artificial Intelligence News
