AI for grant writing: 4 risks nonprofits should understand

In the fast-paced world of nonprofit fundraising, time is often in short supply. Considering that foundation grants alone accounted for 19% of all charitable gifts in 2023, grant writing is essential for securing the resources your nonprofit needs. However, creating grant applications can be time-consuming.

‍

That’s why many nonprofit professionals are turning to artificial intelligence (AI) for help with fundraising efforts. AI promises to streamline the grant writing process by automating tasks, enhancing content, and improving overall efficiency.

‍

While it can be a powerful tool, you need to understand its limitations and the risks involved to set your nonprofit up for success. Let’s explore four key risks when using AI to improve your grant proposals.

1. Over-reliance on AI-generated content

When it comes to nonprofit grant writing, AI can lack the personal touch, unique insights, and deep knowledge about your organization's mission, vision, and values that funders often seek. It can also pull inaccurate or outdated information from online sources, misrepresenting your nonprofit’s needs or impact.

‍

Depending on your exact tools, AI might generate repetitive or generic content, making proposals sound formulaic. Broadly available solutions like ChatGPT generate responses by analyzing vast amounts of internet data. As it attempts to produce helpful responses, it might inadvertently spit out content that resembles others’ work, raising concerns about plagiarism.

‍

By submitting proposals that sound similar to others, you can actually decrease your chances of standing out among other applicants.

How to mitigate this risk

Your grant proposals must be authentic, capture attention, and position your nonprofit as the best option for the funder.

‍

Especially if your organization has limited resources, AI levels the playing field, allowing you to free up time and compete with bigger nonprofits. However, you’ll still want to add your own creativity. Use AI as a tool for efficiency, not a substitute for your authentic human voice and unique storytelling. For instance, you might use it to:

‍

• Structure your proposals clearly and cohesively.
• Research grantmakers’ giving patterns and speak to their values and priorities.
• Write simple language about your nonprofit’s mission.
• Draft longer sections of proposals based on your notes and supporting materials.
• Edit the content you’ve written for flow and proper grammar.

‍

Always review AI-generated content closely. Have a few sets of eyes on it to ensure originality, creativity, and accuracy.

‍

Remember, no one understands your mission like those who work to push it forward every day. By adding your narrative and authenticity, your passion will shine through in your finalized grant proposal.

2. Data privacy and confidentiality concerns

FreeWill’s guide to AI grant writing explains that generative AI tools that are broadly available (think ChatGPT) are trained on massive sets of public data and pooled user interactions. This means they store and repurpose the information that users provide to improve future responses.

‍

Entering sensitive financial, donor, or beneficiary details can pose some serious risks for your nonprofit, though!

‍

Using AI tools without understanding how they handle your data can lead to privacy breaches or even legal consequences, particularly in sectors where confidentiality is crucial. For example, let’s say your nonprofit provides mental health services. Your organization likely handles medical records, diagnoses, treatment plans, and personal health histories, which all fall under HIPAA regulations.

‍

For these nonprofits, any data breach or misuse of sensitive information can harm people, result in legal consequences, damage their reputation, and cause them to lose trust from clients and donors. 

How to mitigate this risk

Be careful inputting sensitive data, especially when using a publicly available large language model (LLM) system like ChatGPT. Exercise extreme caution to ensure that your tools are secure and that you anonymize your data when necessary. 

‍

For tasks involving confidential data, seek out purpose-built tools, such as Grant Assistant by FreeWill, that are designed specifically for the nonprofit sector and come with strong privacy measures (e.g., end-to-end encryption, data anonymization). Before investing in a solution, check its security procedures. It should comply with data protection laws and have a transparent data use policy.

‍

You might look for a solution that has successfully completed the Service Organization Control (SOC) 2 Type II audit. Developed by the American Institute of Certified Public Accountants (AICPA), this audit is a cybersecurity framework that evaluates a service organization’s internal controls related to security, availability, processing integrity, confidentiality, and privacy.

‍

Arcadia emphasizes that combining secure tools with comprehensive staff training is crucial to effective data governance. Establish clear guidelines and develop prompts for using AI responsibly within your organization to maintain data security. 

3. Bias in AI algorithms

AI systems learn from datasets, and if those datasets contain biases—whether subtle or overt—those biases can be reflected in the AI’s outputs. In other words, AI-generated content can contain language or ideas that favor certain groups and overlook others.

‍

For example, the platform might not fully represent the needs or perspectives of marginalized communities. This bias can lead to grant proposals that leave out important viewpoints or misrepresent the impact of your nonprofit’s work. As a result, your proposal might not effectively communicate the full value of your organization or its mission.

How to mitigate this risk

Again, careful review is essential! Make sure that everything in your grant proposal reflects your nonprofit’s values and accurately reflects the communities you serve.

‍

Incorporate inclusive language and practices into the review process to ensure a fair, accurate portrayal of each group involved. One way to do this is by setting up a panel of internal reviewers with diverse backgrounds and perspectives. This can help identify and address biases or discrepancies, leading to a balanced and inclusive proposal.

4. Lack of customization for specific funders

Some AI tools may not fully capture each funder's specific preferences or priorities. While AI can help streamline your grant writing, it may not always understand the subtle differences that funders look for, such as:

‍

• Tone and language: Funders often have specific wording that resonates with their mission. AI-generated content might not understand that preferred tone, making your proposal feel misaligned.
• Alignment with priorities: Funders may communicate that they focus on certain areas like innovation, community impact, or sustainability. AI might not sufficiently speak to these aspects.
• Formatting and structure: Funders typically require specific formatting and sections in grant proposals. AI might not always follow these guidelines.

‍

Funders look for proposals that reflect their unique priorities, values, and expectations—something that AI might struggle to fully capture. Even if you find the perfect grant to strengthen your technological infrastructure or marketing grant to amplify your reach, a proposal without proper customization may feel generic or disconnected from the funder’s mission.

How to mitigate this risk

Customize every grant application to meet the specific needs and expectations of the funding organization. Even without AI, this is a common mistake nonprofits make—they draft a generic outline and fill in details to save time.

‍

Instead, here are some steps to properly tailor your proposal:

‍

• Research the funder’s priorities. Review the grantmaker’s past funding patterns, key focus areas, and any specific criteria they highlight in their guidelines. This will help you understand the goals they value most, allowing you to add those into AI-generated content.
• Use AI for structuring and brainstorming. Manually adjust the tone and language to match funders’ style preferences.
• Leverage purpose-built AI solutions. All-purpose generative AI solutions can handle parts of the grant writing process but aren’t designed to do so. Instead, opt for AI solutions built for grant writing that account for the language, criteria, and priorities of different funding organizations. In turn, you can craft proposals that resonate.

‍

Image Alt Text: A chart comparing two different types of AI tools for grant writing.

‍

Taking these steps will help strike the right balance between efficiency and customization. The result will be more compelling, funder-focused proposals!

Start searching for your AI grant writing tools

AI can be incredibly helpful when used ethically and strategically. However, using it responsibly requires awareness of the risks and a proactive approach to mitigating those concerns.

‍

As you explore tools, take steps to use AI ethically at your organization. The smartest approach is to use a purpose-built solution designed for nonprofits. When adopting a solution, lay out some ground rules for proper use at your nonprofit. You might create prompts for team-wide use or establish guidelines for reviewing and customizing AI-generated content. 

‍

Above all, remember that technology can't replace your expertise and the human touch that drives your mission! Use AI as a complement to your team’s creativity and passion, not a substitute.

‍