4 Tips for Ministry Leaders
Generative artificial intelligence (AI) tools are now widely adopted and embedded in everyday workflows, including content creation and research. As your ministry evaluates the costs and benefits of AI, this article will highlight some potential risks and limitations. This can help inform clear expectations and guidelines, so you feel confident knowing its use aligns with your ministry’s goals and values.
Below are four critical points for ministry leaders to consider when adopting AI use in ministry.
1. Credibility & Accuracy
AI combines information from multiple sources to produce a single, synthesized response. One problem is that it doesn’t always tell you where the information came from. In fact, some of its sources could be unreliable.
If a generative AI tool receives a prompt it cannot easily answer, it may still produce a response that sounds plausible but is incorrect.
For example, if you ask an online AI tool for a series of quotes from a notable figure, like C.S. Lewis, the generated response may include five quotations—but upon further research, you could discover that one or more of the quotes cannot reliably be attributed to C.S. Lewis.
That’s why it is important to have a human verify all information produced by a generative AI tool. It may include tracing an AI-generated response back to its original source to confirm credible support, as well as independently reviewing and verifying the response before use.
2. Copyright Infringement
Generative AI models are trained on a broad range of data, which may include copyrighted materials. This raises potential concerns about how AI-generated content can be used. These concerns are particularly relevant as ministries may use generative AI tools to create written content or generate images that are made publicly available.
For example, AI-generated images may closely resemble existing copyrighted works or reflect recognizable styles, increasing the risk of unintentional copyright infringement. Because the underlying sources and training data may not be transparent, it is often difficult to determine whether content may infringe on others’ rights.
To help reduce potential copyright concerns, consider using tools such as plagiarism checkers for written content, editing AI-generated material to reflect your ministry’s unique voice and context, and conducting reverse image searches on generated images to identify potential similarities to existing works.
3. Potential Bias
Most AI tools generate responses based on your specific prompt, which means the way a question is framed can significantly influence the answer you receive. Write clear, complete prompts to reduce the chance responses are biased, one-sided or miss important considerations.
This is particularly important in a ministry setting, where decisions may involve pastoral care or sensitive situations that require nuance and sound judgment.
If you use a generative AI tool to assist with content or decision-making, it is important not to rely on a single response. Instead, consider asking for alternative perspectives, identifying potential concerns, or reframing your question to evaluate how the response changes. This type of “stress testing” can help reveal gaps in reasoning or considerations that may have been overlooked in the original prompt.
4. Data Security & Privacy
As generative AI tools become more integrated into ministry operations, they may be used in ways that involve sensitive employee, congregant or donor information. These tools operate by processing user prompts and, in some cases, using those prompts to improve or refine future responses. The information included in your prompt may be transmitted to and stored by a third-party provider and used for model training. This creates potential liability risk if personal or sensitive information is shared without appropriate authorization.
Many generative AI tools allow users to disable training features, which can help prevent the system from using the information you submit to improve or refine the AI model. However, this does not necessarily mean your information is not processed or stored by the provider.
For ministries where generative AI plays a significant role in operations, it may also be worth considering enterprise-level tools, which often include additional security controls and administrative features not typically available in publicly available, consumer-grade versions.
If enterprise-level tools are not used, ministries should not share personally identifiable, confidential, or sensitive information with generative AI tools. Taking these precautions can help reduce the risk of unauthorized disclosure and better protect your ministry from privacy and data security concerns.
Wise Stewardship of AI Tools
Once you understand the risks associated with generative AI, it is important to address them with appropriate guardrails. As an initial step, consider adopting a generative AI policy to help protect against data security concerns, reduce over reliance on AI and ensure AI use aligns with your ministry’s mission and values.
If your ministry uses generative AI, it is important to understand the potential liability risks associated with its use and to have appropriate insurance coverage in place to help mitigate the financial impact of those risks. Potential exposures may include intellectual property claims and data breaches. Ministries should discuss their use of generative AI with their insurance agent to better understand these risks and determine whether their current insurance program appropriately addresses them.
For more information about the use of AI in ministry, visit the Brotherhood Mutual Safety Library.
The information provided is intended to be helpful, but it does not constitute legal advice and is not a substitute for the advice from a licensed attorney in your area.
Cody Lewis is an attorney for Brotherhood Mutual Insurance Company. He collaborates with risk control, underwriting, claims, and outside experts to share how ministries can support their staff and protect their property and programs.
