Three in four survey participants are uneasy about digital twins

People who take part in online surveys are wary of AI copies of themselves. In a Prolific survey among about 3,000 participants in 15 countries, 74% said they feel uncomfortable with a digital twin built from their personal data.
Many had not met the idea before. Almost half had never heard of digital twins and more than a quarter had never heard of synthetic data. A large majority, 93%, agreed that research needs real people to be trustworthy. Most also expect fewer studies for human participants if synthetic data becomes common.
The door is not closed. About half would consider allowing a digital twin under the right conditions. What they ask for is control: the ability to review and correct the twin, a say in how it is used and a share in its ownership.
What it means for fieldwork in the Netherlands
Consent is where this lands for qualitative work. Recordings and transcripts are rich material for AI tools. Participants should hear before the session whether their input is used for anything beyond the study itself. We put that in the recruitment screener and repeat it at the start of each session.

