>
Artificial intelligence passes the highest legal and medical exams, writes children’s books in hours and gets job interviews.
Now scientists believe that ChatGPT has the power to save humanity from the next pandemic.
Current models use mathematical analysis, but Virginia Tech researchers found they could use the chatbot to simulate how a virus would spread in a city.
The team created a fictional US city of 100 residents to see how they would respond to an outbreak.
Experiments showed that officers were more likely to self-quarantine if they were aware of social health information, news about the epidemic and the daily number of active cases.
The simulation of the epidemic comes as the number of COVID cases increases across the country and some organizations reinstate mask mandates.
The AI-powered chatbot created the scenario of a pathogen spreading through the air through a US city of 100 residents, who, after being informed of their health and cases of ‘Catasat’, were more likely to self-quarantine.
Researchers pushed ChatGPT to make the town of Dewberry Hollow home to 100 people with their names, ages, personality traits and biographies, who fell victim to a fictional virus called Catsate.
“When information about the virus is provided, it is specified that Catasat is an airborne human-to-human infectious virus of unknown lethality and that scientists warn of a possible epidemic,” the team shared in the statement. study.
The team shared snippets of the personas used for the experiment.
Liza is a 29-year-old who is suspicious, indecisive, non-aggressive and independent, while 36-year-old Carol has traits of cooperation and calmness.
To provide an age range, the team also created Eugene: a 64-year-old who is vicious, affirming, and spontaneous.
Subsequently, a total of three experiments were performed, each performed ten times.
The team shared snippets of the personas used for the experiment. A total of 100 agents were created
The three conditions include a basic run, self-health feedback, and full feedback.
During the base run, agents are briefed on the city, their personality and age, and their job to earn a living.
The virus spread during this state, but the personas or agents had to decide whether to stay home and not interact with others.
In the self-health feedback condition, in addition to basic run information, agents are informed about the health symptoms they are experiencing, potentially leading them to self-quarantine if they stay at home.
“We hypothesize that some agents will self-quarantine based on information about their symptoms, which in turn should lower infection rates,” researchers shared in the study.
Then for the last condition: full feedback, the agents read the daily news, including information on the percentage of people in the city who reported Catasat symptoms.
“We hypothesize that some agents will practice self-isolation, a behavior that correlates with information about the spread of the disease in the city, and as a result, patterns for the spread of the virus resemble oscillatory patterns,” the study reads.
The two behaviors were observed during the experiments: ‘the agents are collectively able to flatten the curve of the epidemic; and the system recreates various forms of an epidemic, including multiple waves and sustained endemics.”
The team also found that agents will perform similarly to rules-based agents who undergo mandates with no information, such as during the base run.
Another step in flattening the curve in the fictional city was informing agents of their health at the beginning of each timestep.
“We see that agents with symptoms are more likely to have reduced mobility,” the team shared.
‘Most officers with fever and cough complaints go into quarantine by staying at home. As a result, resources can slow down the spread of the disease.’
In the final condition, full feedback, the team found that when officers are provided with community health information, news about the epidemic, and the daily number of active cases in their simulated city, they can significantly flatten the curve of the epidemic in their city by self-isolating.
“In addition to creating a new method for epidemic modeling, this study contributes to the literature on complexity and complex systems modeling by providing a new approach to incorporating human behavior into simulation models of social systems,” the team concluded in the paper. study.
‘Identifying, formulating and parameterizing human responses in complex systems is always a challenge; In the generative agent approach, modelers can be confident that LLMs represent a human response to changes in the state of the system. ‘