Advertising for new generative AI tools seems to be everywhere, touting hundreds of benefits like faster turnaround times and greater efficiency – but at what cost?
AI is rapidly reshaping the face of just about every industry, and law is no different.
This month on the podcast, CPMT attorneys Jeffrey Leonard and Jacob Rich sit down and discuss how the use of AI for legal practice threatens one of the law’s most sacred tenets: the attorney-client privilege.
Attorney-Client Privilege
The attorney-client privilege protects the interests of the client by keeping communication between that client and their attorney confidential. Without it, clients are unable to speak freely and honestly with their attorneys.
That privilege allows clients to entrust their future – and that of their family or business – to their attorneys without fear that their personal information will be at risk.
At CPMT, that kind of trust is the cornerstone of everything we do here. We work hard to protect it at every step.
As Jacob says, “We enjoy attorney-client privilege here at CPMPT. In order to protect that, you cannot use an LLM to assist with a client’s specific details at all.”
How Does AI Endanger Privacy?
When you and I talk about generative AI today, most of the time, we’re talking about large language models, also known as LLMs.
These tools – including ChatGPT, Claude.AI, Gemini, and others – are trained on massive datasets from public sources.
Training is extended when users input data, such that any information they share with the LLM may be stored or influence future outputs.
If sensitive details are entered, they can become part of the model’s accessible database. By doing so, those details become potentially discoverable, and privacy is compromised.
Plus, as Jeffrey says, “Once information is in a LLM’s database, there’s no way of getting it out.”
At CPMT, we avoid these concerns about data privacy by creating our own in-house processes that do not involve AI. This allows us to keep your information safe, so that you can continue to enjoy an attorney-client relationship based on trust and confidentiality.
Human Oversight Still Matters
There are some industries where the use of AI can help with performance and efficiency without major legal risk, but law isn’t one of them.
Our work, when we’re providing legal counsel to our clients, is deeply personal, and its high-stakes nature requires information to remain private.
“The closer you can get to the real world, with things you can actually hold… the safer it is from a digital revolution,” says Jacob.
That’s where human legal professionals like our team at CPMT still shine. Human attorneys are bound by ethical codes and professional responsibility, unlike AI, which operates outside those constraints.
At CPMT, we’ve intentionally chosen to keep our client data separate from any AI-driven platforms.
We provide legal counsel that takes into account all of the myriad complex pieces of your specific story, which AI does not have the ability to understand.
We’re committed to helping you navigate each stage of your life with integrity and compassion – two things that AI could never do.