>Large language models already match median professional performance on the routine tasks that constitute a large fraction of professional billing across legal

>Large language models already match median professional performance on the routine tasks that constitute a large fraction of professional billing across legal research, financial analysis, software engineering, Another article that vomits 1 billion words assuming that agi had been achieved. If you have the audacity to question this claim, what value will be left from this pile of words?