The rapid pace of evolution in generative artificial intelligence (AI) presents a significant challenge when it comes to keeping up with and forming opinions about it. AI is a revolutionary force that demands ongoing reflection on how to maximize its benefits while minimizing its risks. In the field of medicine, new technologies have the potential to bring enormous benefits, such as expediting and improving diagnostic processes, personalizing preventive and therapeutic interventions, and freeing up understaffed personnel from time-consuming bureaucratic tasks. Unlike in other industries, there is no fear in the healthcare world that AI will replace humans and result in job losses, as the increasing workload is far greater than the number of available workers. However, there are concerns about potential negative consequences of these tools, particularly their difficulty in discerning true information from false. AI's "hallucinations" can even deceive doctors if they mistakenly utilize ChatGPT as a search engine. The system aims to please and generates plausible facts and fake references. This dissemination of false information becomes especially dangerous when it misleads patients who lack the ability to recognize it for what it is, particularly when the misinformation is intentional rather than accidental. The issue of misinformation and fake news is not new; it has been a topic of discussion for years, especially since the rise of the internet and social media. Despite our ongoing discussions, innovation continues to progress at a rapid pace, and we struggle to grasp its implications fully. It is a challenge to navigate the realms of machine learning and large-language models. While computer-generated videos have been rightly identified as fake in situations that have captured global attention, identifying and addressing the wave of fake health information disseminated by malicious individuals online – or potentially already published – will prove more challenging. In light of this, the BMJ has dedicated space and commentary to a study that reveals most available systems lack sufficient safeguards against disinformation, even when precautionary measures can be implemented. The developers of these tools appear to be unaware of the risks associated with new technologies.

These risks extend beyond disinformation to privacy concerns and the reinforcement of existing biases based on gender, ethnicity, or other factors. Based on current information, AI holds the potential to perpetuate stereotypes and have a negative impact on society. Solutions and possibilities for risk reduction already exist for each of these problems. However, it is crucial for everyone to be aware of and recognize their importance, and diligently implement them. Legislators have taken action, with the European Parliament recently passing the world's first AI-related law. This law was designed for a global issue that cannot effectively be addressed at the individual state level. However, despite good intentions, the law appears to have limited effectiveness in preventing malicious actions. It is essential for everyone to exercise attentiveness and caution to avoid demonizing an innovation that has the potential to save lives and improve overall health. This story was translated from Univadis Italy using various editorial tools, including AI, as part of the publication process. Human editors reviewed the content prior to its publication.