Six Tips for Using Generative AI

Stephanie Enders, ’02 BA, has learned a lot about generative AI in a short time. As the chief delivery officer for Amii, the Alberta Machine Intelligence Institute in Edmonton, she was tasked with writing the framework to guide employees and clients on how to use generative AI tools and offer guidance for Amii’s industry partners.

“For a long time, artificial intelligence and machine learning have felt like they were for a really specific slice of business — very innovative business, very data-driven business, that took a lot of expertise,” Enders says.

“When these large generative AI tools with simple portals for public use launched, it heightened the excitement around this technology. It made it accessible and tangible as a thing that could be integrated into people’s lives quite quickly.”

The sudden availability of the technology sparked a big demand from organizations and businesses to understand how to leverage the tools and whether jumping right in to use them was the right first step, says Enders. Internally at Amii, even with people who understand AI, there was a need to understand how to apply the new tools to day-to-day work.

Enders offers some tips on using generative AI tools based on her experience.

Understand the limits

Learn about the technology you’re using before you dive in. Be clear about potential pitfalls, such as the danger of sharing personal or proprietary information, which other users will be able to see. “Understand where you feel comfortable, what kind of tasks you’re prompting it to do, the kind of information you’re uploading to the system,” says Enders. She recommends the Amii website as a good place to start. It has a range of blogs, courses and workshops for parents, teachers, businesses and others who want to learn more about AI.

Read the manual

Different generative AI tools have different abilities and risks. Make sure you know what they are. “The folks that are making these tools are doing a pretty good job of explaining them — though you have to consider the source,” Enders says. “Because of the massive interest and influx of questions, companies have become better at documenting how the tool works, how to interact with it and answering the FAQs they’re getting.”

Start slow and in control

Come from a place of curiosity, she suggests. “Approach the technology initially as a fun thing to try, with low stakes.” Put some guiding principles in place so you’re clear about how you want to use it. Keep it simple.

Hang on to the reins

Probably the biggest tip, says Enders, is to remember that generative AI is just a tool. The language component can make it feel more human and intelligent than it is. Don’t be fooled. “Remember that whatever the tool’s output, it’s a starting point for you to continue the evaluation of that work and whether or not it’s useful to you and the task you’re trying to do.”

Talk to your boss

If you want to use the technology in your workplace, as a brainstorming tool, for example, Enders recommends having a conversation with your employer. What would be considered proprietary information in your business or field? What should you not upload to a public tool? It’s best to be absolutely clear about the parameters, she says.

Stick to the rules

On the job, anything produced by generative AI still has to follow the rules and standards expected in your workplace or industry, Enders cautions. And the more regulated the industry, the more intense the vetting process has to be. A chatbot may have been trained, for example, in a jurisdiction with different regulations. “Just because generative AI exists doesn’t mean you throw out all the regulations and processes that have been put in place for your work. All those rules still apply.”