Deepfakes are rapidly weakening trust in online content, with nearly half of consumers now questioning the authenticity of what they see on the internet, according to a new study by iProov, a provider of science-based biometric identity verification solutions.
The research found that 48% of respondents say they doubt “almost everything” they encounter online, pointing to what the company calls a “Great Trust Recession” driven by the increase in AI-generated content and impersonation.
The survey, conducted in the first quarter of 2026 among 2,000 respondents in the United Kingdom and the United States, highlights growing concerns that deepfakes and generative artificial intelligence (AI) are blurring the line between real and fake content.
“AI has blurred the line between real and fake in digital ecosystems, and too many organizations are caught off guard,” said Andrew Bud, founder and CEO of iProov. “Deepfakes are quickly undermining the trust at the heart of the digital economy, ultimately compelling consumers to change their behaviors and, importantly, who they are willing to do business with.”
The erosion of trust is beginning to impact business decisions. The study found that 74% of consumers would switch banks if another provider offered guaranteed protection against deepfake-enabled fraud. Younger consumers are driving this shift, with 41% of respondents aged 25 to 34 saying they would switch immediately, compared with just 14% of those aged 65 and older.
“This study highlights a major shift in consumer sentiment, showing that generative AI is actively undermining the credibility of the institutions people have traditionally relied upon,” Bud said.
Expectations for accountability are also increasing. More than half of respondents, or 52%, believe banks should be legally responsible for losses caused by deepfake-related fraud, signaling growing pressure on financial institutions to strengthen safeguards.
The report also points to changing attitudes toward digital public services. While more services are moving online, 55% of respondents said they would be more likely to use government platforms if secure biometric login options were available. When given a choice, 43% prefer a mobile-based face scan for verification, compared with 30% who favor in-person appointments.
To illustrate how convincing deepfakes have become, iProov introduced an interactive test called “Find the Fake,” where users try to identify a single AI-generated profile among six realistic social media accounts. The exercise underscores how difficult it has become to distinguish real identities from synthetic ones in everyday digital interactions.