Safety and Trust Checklist Before Using AI in Your Course

{"raw": "

Before you commit to using an AI tool for your course, verify that your content and students’ data will be protected.

nn

The Privacy Questions You Must Ask

n

Visit the tool’s privacy policy and find answers to these specific questions. First, will the company use your content or student data to train their AI models? Second, do they encrypt data in transit and at rest? Third, where do they store your data—which country? Fourth, can you delete all your content and data whenever you want? Fifth, do they share data with third parties? If you can’t find clear answers to these questions, don’t use the tool. Most educational AI tools allow you to opt out of training data use.

nn

The Student Privacy Question

n

If your students will interact with the AI tool, verify FERPA compliance (if you’re in the US) or equivalent privacy laws in your region. Check whether the tool has specific protections for student data. Many tools can’t be used with student data due to privacy requirements. Make sure you check before you build your course around the tool.

nn

The Staying Power Question

n

AI tools get bought, shut down, or change features constantly. Before investing time in learning a tool, check: Has the company been around for at least a year? Do they have paying customers or are they purely venture-funded? What’s their history of product stability? Is there an exit plan if the tool disappears? Good rule: Use newer AI tools only for content that you can easily recreate or export. Use established tools for content you plan to keep long-term.

nn

The Data Export Question

n

Can you export all your content from the tool in a format you can use elsewhere? Never build your course in a tool that doesn’t allow you to get your data back in a usable format. This is crucial for independence. If a tool disappears, you need to be able to move your content elsewhere.

nn

The Transparency Question

n

Does the company clearly explain how their tool works, its limitations, and its failure modes? Companies that hide how their AI works raise red flags. Companies that openly discuss limitations build trust. Choose tools from companies that educate you about AI rather than overselling capabilities.

nn

The Bias and Accuracy Question

n

Does the company publish information about their model’s bias, accuracy rates, and limitations? Good companies test for these issues publicly. If they claim perfect accuracy or never mention limitations, that’s a warning sign. No AI tool is perfect—trustworthy companies talk honestly about where they fall short.

nn

Rule: You’re responsible for content in your course. Use tools only from companies you trust with student data and your intellectual property.

"}