On Tuesday, lawyers for OpenAI walked into a Chicago federal court and made an argument that would have seemed unthinkable two years ago. Their flagship product, the chatbot that has reshaped global conversations about artificial intelligence and pulled in tens of billions of dollars in investment, was, they insisted, not really doing anything at all.
The case was brought by Nippon Life Insurance, which accused ChatGPT of effectively practising law without a licence. A self-represented litigant had allegedly used the chatbot to generate a torrent of legal filings, many of them meritless, in a dispute that had already been settled. Nippon Life argued OpenAI should be held responsible for what amounts to unlicensed legal practice — a serious accusation in a profession that guards its credentials fiercely.
OpenAI's defence rested on one extraordinary sentence buried in its motion to dismiss. ChatGPT, the company told the judge, "is not a person and neither has nor uses any degree of legal knowledge or skill."
Read that twice. This is the same company whose chief executive has spent two years telling anyone who will listen that their AI is approaching graduate-level reasoning. The same company whose latest models are pitched to law firms, accountancy practices and consultancies as productivity multipliers. Now, in court, that product is just a glorified autocomplete.
