Imagine you’re facing a serious legal issue, like being accused of murder or drug charges. You might think to turn to an AI assistant like ChatGPT for advice, but experts warn that this could be a recipe for disaster.

According to OpenAI CEO Sam Altman, if you’ve asked ChatGPT about sensitive legal topics, you may have inadvertently created evidence that can be used against you in court. This is because ChatGPT conversations are not protected by confidentiality agreements like human lawyers’ discussions are.

“Right now, if you talk to a therapist or a lawyer or a doctor… there’s legal privilege for it,” Altman explained during a recent conversation. “There’s doctor-patient confidentiality, there’s legal confidentiality.” However, he acknowledged that OpenAI has yet to figure out how to provide similar protections when users interact with ChatGPT.

Lawyers like Jessee Bundy of the Creative Counsel Law firm have been warning about this issue for over a year. “If you’re pasting in contracts, asking legal questions, or asking [the chatbot] for strategy, you’re not getting legal advice,” she tweeted. “You’re generating discoverable evidence. No attorney-client privilege. No confidentiality.”

Bundy’s warnings were sparked by recent conversations with AI defenders who downplayed the risks of using ChatGPT for legal purposes. However, Bundy is adamant that users are creating a false sense of security when they interact with ChatGPT.

What Makes ChatGPT Different from Human Lawyers?

So what makes ChatGPT different from human lawyers? The main difference lies in accountability and responsibility. A licensed professional has the expertise to understand your specific situation, goals, and jurisdiction, and is accountable for their advice.

On the other hand, ChatGPT is a language model that generates words based on patterns it’s learned from data. While it can provide responses that sound right, it doesn’t have personal knowledge of your case or circumstances, and is not responsible if its advice turns out to be incorrect.

This means that any “advice” you receive from ChatGPT should be treated with a healthy dose of skepticism, according to Bundy. She warns that users are essentially playing “legal Mad Libs” when they rely on AI for sensitive legal matters.

Inspired by

• https://futurism.com/chatgpt-legal-questions-court

Sources

• theconversation.com
• kasparlugay.com
• techcrunch.com
• openai.com