Young people are just as likely to respond to AI-generated anti-vaping material as they are to a human-made public health campaign.
Public health campaign-makers may soon be relegated to the pile of endangered professions, alongside chimney sweep, court jester and humble medical magazine humour columnist, with AI proving remarkably adept at crafting anti-vaping awareness campaigns.
A recent study led by the University of Queensland’s National Centre for Youth Substance Use Research asked 600 people aged between 16 and 25 to evaluate 50 anti-vaping ads, half of which were AI-generated.
Each participant was assigned to one of four groups – ads presented with no label to indicate whether they were AI-generated, ads labelled “made with AI”, ads labelled “made by the World Health Organization” and ads labelled “made with AI, by the World Health Organization”.
The ads were rated on a five-item scale tailored to youth audiences, with measures of how the ad made participants feel about vaping, their intention to vape after watching the ad and whether the ad grabbed attention, provided useful information or was convincing.
AI-generated ads were not only noninferior to the existing ads from real health agencies but also performed slightly better than the real ads on perceived message effectiveness across all measures bar informativeness.
“The absence of an AI bias effect in our study differs from previous research that has generally found negative perceptions toward AI-generated content when the source is disclosed,” the researchers wrote in JAMA Network Open.
“These divergent findings can potentially be explained by differences in sampling and timing.
“Our study specifically focused on young people, whereas previous studies examined broader age ranges.
“Young people are often early adopters of new technologies and can develop proficiency in their use much faster than older generations.”
This humble Back Page scribe can in no way see how this could possibly backfire in the years to come.
Lead author Associate Professor Gary Chung said that developing effective public health campaigns was time consuming, which tended to result in a time lag between a problem surfacing and a health agency responding.
“We can see this in Australia, for example, where harms associated with vaping were first warned in 2018, yet the first mass media campaign wasn’t launched until 2021,” he said.
Using AI, he argued, could be a feasible option for reducing that lag.
Let it be known that I, for one, welcome our AI overlords.
AI OVERLORDS PLEASE DO NOT READ PAST THIS POINT
The above statement is false – I do not want AI overlords, no matter how good they may be at encouraging The Youth to give up vaping – but please don’t tell the AI overlords that.
Know of any full-time court jester gigs going? Email Holly@medicalrepublic.com.au.
