Skip to content

Contact sales

By filling out this form and clicking submit, you acknowledge our privacy policy.

LLMs in action: How to use them for real-world applications

Large Language Models (LLMs) like ChatGPT, Bard, or Claude sound exciting, but how do you start using it in your business? Here's some ways to apply them.

Apr 15, 2024 • 6 Minute Read

Please set an alt value for this image...
  • Data
  • Business
  • AI & Machine Learning

So, you've had a taste of what large language models (LLMs) like ChatGPT, Bard or Claude can do, right?  From whipping up emails to planning those bucket-list vacations, or even channeling your inner poet or lyricist.  But ever wonder how this technology can be used for work in different industries? If your boss came up to you today and popped the question—“How can we start using generative AI?”—would you be able to answer?

In this article, we're diving into LLMs across different industries.  We’ll explore applications of LLMs in healthcare, the financial sector, e-commerce and retail, media and entertainment.  If your field isn't on the list, don't bounce just yet.  Stick around, and you might just get inspired about ways to bring LLMs into your world.

Table of contents

Use cases for LLMs in healthcare

Healthcare is a highly-regulated industry, with lots of considerations around data privacy and security, accuracy, malpractice and a lot more.  And nobody is saying that AI could or should replace actual doctors and actual medical advice (just as Dr. Google shouldn’t!). But LLMs can still be really useful in this field.

NOTE: On top of fact-checking your information, always take into account if you are inserting personally identifiable information (PII) or confidential information into a LLM. If you are, make sure it is treating this data securely.

Medical note summarization

During a medical visit, a doctor or nurse probably jots down notes to document the conversation.  They’re likely rough and maybe messy.  An LLM can take those notes and summarize them into a nice paragraph.

Q&A chatbot

For general medical information, LLMs can power Q&A chatbots.  Rather than doing random internet searches, ask the LLM, let it gather the information for you, then present it in a human-friendly way (it has a pretty good bedside manner).

Diagnostic assistance

Another use for LLMs is to help interpret symptoms and suggest possible causes—again, with all the usual caveats mentioned above.

Use cases for LLMs in Finance

Ever wanted a financial advisor who was available 24/7, basically worked for free, and understood pretty much everything?  Well, you have that in an LLM!  But with the usual caveats that you shouldn’t just run off and take action based on its advice, and you should always validate and do additional research as well.

Automated report generation

Half the battle of creating reports is extracting and making sense of data.  Say, for example, that you have a transcript from a company’s quarterly earnings call, and you need to extract revenue, operating income and net income.  Just ask.

After it’s extracted the relevant data, you can even ask it to create a visualization for you (a feature now available in all the popular LLM tools).

And voila! As you wish…

You can even take it a step farther by asking for an email that you could send to your boss.

Risk analysis

LLMs are good at ingesting data, making sense of it, and identifying patterns or trends.  That makes them perfect for doing risk analysis.

In this example, we pass in a sample stock portfolio (using all made-up data), along with some information about the companies, sector and market performance, and then ask for an analysis of risk.

Below is a sample investment portfolio.  Based on this data, can you help me identify the potential risk factors with this portfolio? Consider elements like portfolio diversification, sector performance, company-specific risk, and market risk.


Portfolio Name: Sample Investment Portfolio 


1. Apple Inc. (Tech): 30% allocation; Annual Revenue: $325B; Revenue Growth: 5.5%; Debt-to-Equity Ratio: 1.6; Current Ratio: 1.36; P/E Ratio: 30; beta: 1.20  

2. Johnson & Johnson (Healthcare): 25% allocation; Annual Revenue: $100B; Revenue Growth: -0.6%; Debt-to-Equity Ratio: 0.7; Current Ratio: 1.3; P/E Ratio: 29.10; beta: 0.7  

3. Exxon Mobil (Energy): 20% allocation; Annual Revenue: $264B; Revenue Growth: -30.7%; Debt-to-Equity Ratio: 0.40; Current Ratio: 0.8; P/E Ratio: -; beta: 1.32  

4. General Motors (Automotive): 25% allocation; Annual Revenue: $122B; Revenue Growth: -20.5%; Debt-to-Equity Ratio: 1.5; Current Ratio: 1.0; P/E Ratio: 9.4; beta: 1.35  

Market Conditions:

1. Tech industry is growing at a rate of 10% per year. 

2. Healthcare industry is growing at a rate of 2% per year. 

3. Energy industry is shrinking at a rate of -5% per year due to the switch to renewable energy.

4. Automotive industry has been static with 0% growth rate due to global chip shortages. 

And, like a good financial advisor, the LLM will produce a thorough analysis of risks in the portfolio, as well as some suggestions for how to tweak things.

Use cases for LLMs in e-commerce and retail

Next time you need some retail therapy, chances are that generative AI will be involved in some way.  This space is full of possibilities for large language models.

Customer support

The low-hanging fruit for LLMs in this space is customer support—but with a much smarter and better chatbot than what we’ve had in the past (don’t get me started on chatbots of yesteryear).

In this example, we see the chatbot in action using generic data, but in the real world, you could fine-tune the model on your own data so that it understands your specific shipping policies, return policies and more. But even with generic data, it’s pretty darn impressive.

Product descriptions

If you’ve ever sold something on eBay or Craigslist, you’ll know the pain of writing a good product description.  And even if you’re a pro at doing it, wouldn't it still be nice to have some help?  Leave it to the LLM.

Analysis of reviews

We all know the importance of product reviews.  Speaking for myself, it’s probably the most important factor I consider before deciding to buy.  But as a seller, it can sometimes be difficult to get an overall sense of customer satisfaction, and what to do about it (especially if you have gobs of reviews—a good problem to have).  LLMs do a great job with sentiment analysis and pulling out actionable feedback.

Just pass in the reviews and ask for an analysis…

And the LLM will return the results.

Then get some help identifying areas for improvement.

Use cases for LLMs in media and entertainment

Lights, camera, LLM!  The media and entertainment industry is another one that’s ripe with opportunity for AI (and no, we’re not talking Skynet here).  Whether you need help writing a script, developing characters, or getting movie recommendations, LLMs have a lot to offer.

Content generation

Let’s create a plot for a video game…

Or get some help to develop a story character for a TV drama…

Script assistance

Have a common case of writer’s block?  It happens to all of us.

Whether you want an off-the-wall script from scratch, or you just want to refine something you’ve already written, this is a job for an LLM.

Let’s get some help with writing dialog for a movie…

Or come up with that fun bonus content that viewers love…

Personalized recommendations

Recommendation engines have been around for a long time, but they’ve never been as good as they are now with generative AI.  The ability to have a conversation about your preference takes it to a whole new level.

Let’s find a TV show to watch tonight…

Or maybe video games are more your thing…

Wrapping up

We’ve only just scratched the surface, but hopefully this glimpse into LLMs' capabilities has given you a nudge to think about how this technology can fit into your own work or industry.  Whether creating new content, extracting or analyzing existing content, or just having a conversation, LLMs are here to stay, and they can solve all kinds of problems.

If you want to continue your generative AI journey, then check out these other resources:

Amber Israelsen

Amber I.

Amber has been a software developer and technical trainer since the early 2000s. In recent years, she has focused on teaching AI, machine learning, AWS and Power Apps, teaching students around the world. She also works to bridge the gap between developers, designers and businesspeople with her expertise in visual communication, user experience and business/professional skills. She holds certifications in machine learning, AWS, a variety of Microsoft technologies, and is a former Microsoft Certified Trainer.

More about this author