Illustration with the letters AI in the center of a white circle outlined in blue. Branching out from that circle are many blue lines and white circles outlined in blue. In the background are darker circles.

Generative AI (GenAI) is everywhere, and it’s impacting ecommerce, design, and content creation. Now, manufacturers are getting in on the action.

They are understandably cautious—yet there’s also excitement and curiosity. Large language models (LLMs) like ChatGPT open new avenues to innovate in ways previously unimagined.

This builds on the AI advancements we already see in PIM and DAM. For example, some PIM platforms can apply AI to create specs and product descriptions, and intelligently suggest and apply data field mappings for both individual SKUs and even entire product lines. In DAM, it can automate metadata creation and image tagging, recognize product context and apply the appropriate taxonomy, identify safe spaces in images for text, and more.

GenAI takes these capabilities a big step forward. Some PIM and DAM platforms even offer connectors to LLMs like ChatGPT, providing a golden chance to try it out.

Five practical GenAI experiments for manufacturers

Despite concerns surrounding GenAI, its adoption is undeniable. A 2023 study by business.com revealed that 57% of American workers have experimented with ChatGPT, finding it a tool that helps them work “smarter, not harder.” Furthermore, 80% of Fortune 500 companies have implemented GenAI in some capacity.

As GenAI gains traction in industries like manufacturing, it’s time to look at some hands-on experiments you can try. Here are five examples.

Icon of a white, counterclockwise circular arrow enclosed in a white circle above a white arrow pointing right. This is on a black background outlined in blue.

1. Iterate on product design

GenAI can be a helpful tool for iterating on product design. By feeding it images and asking for variations, it can provide both engineering insights and stylistic inspiration.

Let’s say you’re working on a new line of excavators for different climates.

You could upload an image of an existing design and say, “Show me this excavator adapted for use in tropical outdoor conditions.” The system will then generate a new image to meet your request. I know this example sounds elementary but try it with an example from your industry. GenAI has access to so much data out there. It can “think” and apply concepts in a new way that might turbocharge your next brainstorming activity.

You can continue by asking ChatGPT to explain what was changed and why. Use that information to continue iterating on the design. You may even want to input user reviews and ask ChatGPT to provide 20 ways to improve the product based on that feedback.

What GenAI returns is just the beginning, not the final result. The goal is to get fresh ideas and insight quickly to guide your design journey.

White icon of stacked squares on a black background outlined in blue

2. Ask for an exploded parts diagram

This is another example of rapid inspiration. You can use GenAI to create exploded parts diagrams of machinery, even if the product is a new concept.

For example, if you ask for an exploded parts diagram of a vacuum cleaner, ChatGPT can generate a wealth of information that you’d otherwise spend hours or days tracking down. While not always perfect, the result can be invaluable for reverse engineering or creative brainstorming.

White icon of a checklist on a black background outlined in blue

3. Create a safety checklist 

Ask ChatGPT to create a safety checklist for a particular product type or specific part number. This is a great way to jump ahead when building a checklist for the very first time. You could also input an existing list and ask ChatGPT to suggest items you may have missed.

This works for other aspects of your product, too, such as any applied software testing. LLMs have access to a massive amount of software testing protocols.

White icon of three boxes connected by a line leading to an arrow pointing right. This is on a black background outlined in blue.

4. Summarize information

GenAI can be fantastic for turning detailed discussions into clear, concise summaries. Let’s say you just wrapped up a technical meeting with your engineering team and you need to share key points with members of customer service.

Ask ChatGPT to provide a summary of the main points for the business reader using simplified language. Ask it to tailor the summary specifically for your customer service team. Or use it to create a list of action items or potential next best steps based on what was discussed.

In 2024, AI notetaking is nothing new. But remembering to prompt the AI to consolidate notes for a particular audience, such as customer service, can go the extra mile towards ensuring better communication.

White icon of a screen with a person on it. This is on a black background outlined in blue.

5. Create video avatars for support

Creating video avatars for customer support is on the aspirational side. However, there are platforms already offering this today, and the technology is improving at a staggering rate. Ntara has a client using this for internal support purposes with great results. A user can input a customer service script, and the program creates a digital representative that can interact with customers/users, answering their queries in a more engaging way than traditional text-based chatbots.

For example, you could start small by creating a video avatar to handle one aspect of your operations, such as warranty queries. Choose a GenAI platform that supports avatar creation, input a script focused on common warranty questions, and test how it enhances customer interactions and saves you time.

Skeptical? Check out Synthesia for yourself. You can even create a test video for free to see how far the technology has come along.

Cover your bases

These experiments are low-risk ways to try out GenAI in manufacturing. But as adoption grows, consider putting certain boundaries in place for accuracy, quality, and security.

  • Accuracy
    Is the information factual and correct? LLMs often fabricate details, so a knowledgeable human should always review and verify.
  • Quality
    Does it make sense and meet your standards? Learn which prompting techniques are most effective and have a skilled individual review aspects like tone, voice, and messaging.
  • Security
    Are your interactions secure? Strengthen your integrations and internal policies.

Some manufacturers might see the need for such guardrails as a deterrent to experimenting with GenAI—but the opposite is true. By setting these processes up now and honing your GenAI skills, you’ll have a leg up as the technology evolves and becomes more widely adopted by competitors.

It’s time to test the GenAI waters

For manufacturing, fully adopting GenAI will require a coordinated effort between IT, security, product design, and the executive leaders who will chart a course for its strategic use in the business.

In the meantime, test the waters. Start small with some of these ideas using safe, non-confidential data. This is your chance to get comfortable creating prompts and exploring how GenAI can help you accelerate manufacturing processes and spark new ideas.

Want to stay in the loop on future Ntara content? Subscribe to our blog.