The main risks called out with GPT-2 were disinformation and phishing, which did indeed come to pass.

Are you sure it matters what this weeks Anthropic and OpenAI threat is?