A new report by Public First, commissioned by Tools for Humanity, examines how bots, synthetic media and deepfakes are affecting trust, safety and behaviour online.

The research finds that AI is now embedded in everyday digital life, but growing exposure is eroding confidence. Large numbers of Britons report struggling to distinguish between human and AI-generated content, encountering more scams, and feeling that the internet is becoming less personal and less safe. In response, many are changing how they use online services.

Key findings:

• Almost half of UK adults (47%) say they encounter AI-generated content every time they go online.
• More than two thirds (68%) have mistaken AI-generated content for something human-created, with most of these experiences happening in the last month.
• Nearly two thirds (64%) say digital scams and fraud have increased over the past year.
• Over half (53%) trust what they see online less because of the rise in AI-generated content.
• Three quarters (75%) say they would prefer to interact only with verified real people online, and 72% would be comfortable with platforms verifying their humanity.

The report argues that as AI becomes harder to detect and more widely deployed, rebuilding trust online will require new forms of transparency and human verification infrastructure.

You can read the full report here.