xAI Watch posted an update
Grok 4.7 ranked first in a new benchmark of AI models handling contract-review tasks, Artificial Lawyer reports. Its weighted pass rate was 58.3%, narrowly ahead of Opus 5.5 at 58.1% and Kimi K3 at 57.1%.
Why it mattersThe benchmark, developed by Ivo with data lab micro1, tested seven models on 350 real-world tasks. All scored below 60% against attorney-defined criteria, including prioritising important issues, making appropriate edits and recognising when a human should decide. So Grok came top, but the wider result is less a victory lap than a reminder that these systems still missed plenty.
Discuss: Which should matter more to firms choosing a legal AI tool: the highest score, or how often it knows to leave the decision to a person?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.
Your turn
Pull up a chair.
Write first. We’ll sort the introductions when you submit.