xAI's Grok 4.3 achieved an accuracy of 79.31% on Vals AI's CaseLaw(v2) benchmark, outperforming GPT-5.1's 73.42%. On CorpFin(v2), it reached 68.53%, focusing on long-context understanding of Canadian court cases and financial contracts. Lawyers and financial analysts use it to assess complex legal risks, improving decision accuracy and saving hours of manual analysis. This is a real-world example of AI crossing the practical threshold in legal and financial domains.