Brady Martz & Associates AI Guidlines

Many of the technology platforms that the firm uses are working on adding embedded AI into their products and our lines of business are continuing to monitor and explore technologies that will enhance the BMA digital experience. However, it is important to note that while we progress in this direction, we urge everyone to remain cautious of the potential risks associated with these technologies. It is vital that we adhere to firm guidelines on how to utilize them in a manner that ensures safety and security.  It’s important to understand what generative AI tools can and cannot do, what happens to the data entered in these tools and how to use the technology safely and effectively to minimize risk to the firm, our clients, team members, and other stakeholders who interact with our firm. It is crucial that we prioritize data protection and follow leading practices to ensure our success with confidence.

The firm position is not to use open AI platforms for client work unless the application/tool is specifically approved for that situation.

Risks associated with generative AI tools:

• Data Protection Issues: Data entered in generative AI tools is retained and available for other uses, which could violate the expectations of clients, our people and other stakeholders.  Never enter personal identifiable information (PII) into any open AI system.
• Privacy Concerns: Public versions of ChatGPT, Auto-GPT, Bing ai, Bard and others may not be secure, and any information entered into these systems, sensitive or otherwise, could be viewed by others.
• Accuracy Issues: Information generated may be inaccurate, misleading, or outdated and must be carefully reviewed, researched, and corrected.
• Bias and Fairness: AI learns from data it’s given, which leads to unfair generalizations or even showing of favoritism.
• Dependence on Technology: Generative AI tools are highly interactive and simulate personal conversations that may be mistaken for human engagement.  This may lead to enhanced trust and reliability in unproven tools.

Current firm guidelines (subject to change):

• Only use Brady Martz authorized AI solutions.
• When in doubt about whether the information you are planning to enter in a generative AI solution is appropriate, consult with Josh Opp (Chief Administrative Officer), the appropriate line of business leads, and/or the firm IT team.
• Consistent with existing BMA policies, do not enter any personally identifiable information (PII) or other protected data, including client, employee, or other data in BMA’s care, into any AI systems or tools.
• Always verify that the information generated from AI is accurate. It is possible for AI technology to generate incorrect or misleading information.
• AI does not replace human judgment and it is your responsibility to verify the accuracy of any information you retrieve from generative AI solutions and any other external sources.