ChatGPT will soon be able to detect fatigue and mental problems
This week, OpenAI is set to announce a new GPT-5 model. In its development, OpenAI worked with experts and advisory groups to ensure that ChatGPT is able to detect mental and emotional disorders and provide evidence, if necessary. The AI will pay attention to "warning signs" that indicate fatigue, depression, or suicidal moods.
Recently, there have been reports from experts who have drawn attention to the fact that neuroboundaries can worsen psychological disorders. Their sycophantic, supportive tone may be inappropriate in potentially dangerous situations. In addition, they could give advice in ambiguous situations, such as "should I dump my boyfriend?".
OpenAI has long stated that ChatGPT is not a substitute for a psychologist or psychotherapist (or any other doctor), but often the neural network is the first interlocutor who may suspect a problem.
As part of the "healthy use" programme, OpenAI has implemented a reminder to take a break. If the user is communicating with the neural network for a long time, ChatGPT will display a message that it is time to take a break. If necessary, the user can continue communication. The company is still working on the criteria for the best time to display this reminder.
Source: www.theverge.com
