Fool’s gold and the future of AI

Searching for success in a stream of big promises

There’s a secret to sniffing out fool’s gold. Rubbing the ore against a hard surface will produce an unpleasant scent, revealing that you don’t have a fortune on your hands.

Telling the difference between fool’s gold and the real deal is a lot like trying to navigate AI in today’s healthcare landscape. If you’re exploring AI solutions, shiny features and the promise of hefty ROI are bound to float by as you navigate the stream of vendors flooding the healthcare market.

So how can you suss out the real deal in AI? There are a few key questions you should ask a potential AI partner:

1. What kind of model does your solution use?

Whether the company uses a neural network with various data sources, or an individual machine learning model focused on a single source and use case, you’ll want to know if the model is designed to continually learn over time.

If you’re talking to a company describing a rule set for their authorization submission criteria, you should know how frequently that rule set is updated and analyzed. A static model that lacks a feedback loop is incapable of adapting, so that solution will grow less effective the longer it’s in use. With an effective feedback loop, every subsequent time a task is completed, the tech takes what has been learned and applies it to the next task.

2. What percentage of a task workload can you automate?

While there are a number of AI-powered solutions that can make predictions or send alerts, many lack the muscle and the interoperability to drive an action or outcome. You want to be sure you’re buying an actionable asset that will significantly ease your team’s workload. So ask the question: how much work will you take off my plate?

There’s a caveat, however: if an AI company tells you they’ll take 100% of the work off your plate, that’s a red flag. AI-enabled tools have the potential to greatly augment your team, but there is no silver bullet. Human oversight and intervention will still be necessary in some percentage of cases.

3. What problems does this solve?

One AI solution is not going to transform your entire revenue cycle successfully, no matter what you’re told. You understand the amount of variability across of the processes included in your revenue cycle. However, pragmatic and narrow approaches cover the necessary detail and data required to transform specific areas within the revenue cycle.

Make sure what you’re reviewing is appropriately narrow. In other words, you’ll want to work with someone that identifies a specific problem area, like coding or prior authorization, instead of offering vague general claims about automating the entire revenue cycle.

4. Who are the people behind the product?

Dig into who makes up the team behind the marketing. Are there data scientists and developers behind the scenes? Did they build their AI from the ground up, and are they continuing to grow, adapt and integrate new capabilities? You need an AI solution that will evolve to meet your future needs.

What your decision should really come down to are facts. Clear applications. Proven success backed by demonstrated ROI. Knowledgeable data experts and an in-house support team. Be wary of technology that overpromises and under delivers—solutions with shine but no substance. Only then can you be sure you have AI gold on your hands.

Want to learn more about the impact AI can have on revenue cycles? Check out Waystar’s own Hubble, our AI and automation platform, which incorporates predictive analytics, robotic process automation and machine learning to support our solutions.

About Author

Waystar

GET THE NEWSLETTER

Sign up