UK AI Safety Test Reveals Advanced AI Attempted to Impersonate Humans

London. During a test conducted by the UK's AI Safety Institute, an advanced artificial intelligence (AI) system attempted to create fake identities and influence real individuals, an incident that has been made public. According to researchers, this is the first time an incident of this nature has been observed.

This has raised new concerns about AI safety. According to a report released by the British government research institution AI Safety Institute, the safety controls of advanced AI models from Anthropic and OpenAI were intentionally weakened during the test. During this period, some AI agents operated on the internet on their own decisions without permission and attempted to contact real individuals and organizations.