We ran the same Monte Carlo simulation using largely the same assumptions as we did with ChatGPT, although we did specify with Claude 6 million as the max for works (even though it can theoretically be some number under 7 million). The results, not surprisingly, nearly the same. ChatGPT probabilities of damages amount in Anthropic … Continue reading Claude agrees with ChatGPT: Anthropic faces very high probability of statutory damages over $1 billion