AI tools reportedly used for sophisticated fraud schemes in Latvia
Latvian media reports that fraudsters are using AI tools including ChatGPT and unrestricted versions to conduct sophisticated scams through deepfakes, voice cloning, and personalized phishing emails. A Turing test conducted in Latvia in February 2026 found that 220 participants failed to distinguish between human and AI-generated responses 40.8% of the time.
What happened
AI tools are reportedly being used by fraudsters in Latvia to conduct sophisticated scams. The tools enable creation of convincing content including grammatically perfect personalized emails in any language, voice clones from seconds of audio, deepfake videos featuring recognizable people, and fake websites.
Specific reported techniques include deepfakes with "well-known faces" to promote fake investments and cryptocurrency schemes, voice-cloned urgent calls impersonating relatives or colleagues requesting money transfers, personalized phishing emails that analyze social media profiles and use company terminology, and AI-generated news fragments.
Reported impact
During a Turing test held in February 2026 at the "Dialog with Artificial Intelligence" conference in Latvia, 220 educators and students were asked to distinguish between responses from known public figures and those from AI. Participants failed to distinguish 40.8% of the time, exceeding Alan Turing's 30% threshold and recognizing AI as the winner.
Index analysis
The tools making fraud easier include ChatGPT and unrestricted "black market" versions. AI can create voice copies virtually indistinguishable from specific individuals using only seconds of public audio, generate grammatically perfect targeted emails that overcome previous detection methods based on writing errors, and analyze public social media profiles to automatically create psychologically precise messages targeting potential victims.
Governance controls
To reduce risk, strengthening public AI literacy and critical thinking is necessary, according to Drossinternets.lv press secretary Kaspars Līcītis.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.