Corti Beats OpenAI by 25 Percent in Medical Coding

In this article, you’ll discover:

• How Corti’s new tool beat OpenAI by 25 percent in tests.
• The real cost of medical errors and how AI helps.
• Why an expert system must think and not just label.
• How this works in the United States without extra training.

When it comes to healthcare, getting the details right is very important. Recently, a company called Corti released a new platform named Symphony. This platform handles medical coding. In tests, it beat big names like OpenAI, Anthropic, Google, and Amazon. It showed a 25 percent accuracy edge over these famous models.

Medical coding is not just simple data entry. It is the core of how hospitals run. Andreas Cleve, CEO of Corti, explains this well. He said, “Medical coding has been treated as a back-office cost center for decades. It isn’t, it’s the data layer that healthcare runs on. Getting it right changes what health systems can see, decide, and do.”

Because this is a complex job, AI often struggles. Lars Maaløe, CTO of Corti, shared why their tool is different. He stated, “Most AI systems fall short in medical coding because they treat it as labeling, not reasoning. Correct coding depends on evidence, context, hierarchy, and guideline interpretation. We built Symphony for Medical Coding to follow the same decision process expert coders use, and that is why the performance gap is so meaningful.”

Medical experts are already noticing this big step forward. Steve West from Healthliant Ventures added, “The methodology behind Code Like Humans is the most promising approach to medical coding we’ve seen. We’ve been co-developing with Corti because we believe specialized AI infrastructure is how this problem gets solved, and we’re excited to see it move into production.”

The High Cost of Coding Errors

Medical coding turns a doctor’s notes into organized data. This data is used for billing and tracking diseases. When errors happen, they cost a lot of money. But the human cost is even worse.

In one study using Danish patient data, Corti found three times as many suicide attempts as human coders had recorded. The information was in the notes. However, human coders working under tight deadlines missed them. When details are lost, hospitals cannot offer the right help.

A Tool That Thinks Deeply

The American coding system alone has 70,000 different diagnosis codes. Guidelines change all the time. This makes the job a huge reasoning task.

To fix this, Corti ran a massive study with almost six million patient visits. This led to a framework called Code Like Humans. It mirrors how a real expert coder thinks. It checks evidence and follows strict rules. Symphony uses this setup to provide top-quality results at a lower cost.

Working Across Borders

Most AI tools need special training for different places. Symphony does not. It is the first system built to work in both the United States and European countries right away.

It can handle different coding systems without needing local updates. This means hospitals everywhere can use it to track health trends accurately and safely. It leaves a clear trail so auditors can easily check the assigned codes.