AI in Action: Unprecedented Innovations in Taxation, Medicine, and Pathology

So, what are those more exciting use cases?

The Internal Revenue Service (IRS) plans to use artificial intelligence (AI) to audit 75 of the largest partnerships in the United States, each with an average of $10 billion in assets, by the end of September. The AI will help identify patterns and trends that were previously undetectable, and the IRS will also assign dozens of Revenue Officers to the same task. The program will target individuals with reported incomes above $1 million who have more than $250,000 in recognized tax debt and will expand its focus on digital assets. The IRS began experimenting with AI for general income tax accounting in 2021 and introduced voice and chatbots to assist taxpayers.

An AI tool developed by researchers at the University of Rochester can assess the severity of Parkinson’s disease and outperform primary care physicians. Patients can participate from home by tapping their fingers in front of a webcam, and the AI model generates computational metrics to assess the severity of tremors. While expert neurologists outperformed the AI model, it significantly outperformed primary care physicians, highlighting its potential in primary care settings and improving healthcare access and equity.

An AI-driven brain implant has allowed a stroke survivor with locked-in syndrome to speak again after nearly 20 years. The implant uses AI to decode the woman’s brain signals while she tries to speak and then synthesizes speech and facial expressions using a computer-generated avatar. The researchers behind the project claim that this is the first successful use of neurotechnology to synthesize speech and facial expressions from brain signals in a patient with locked-in syndrome.

Microsoft is partnering with digital pathology provider Paige to create the world’s largest image-based AI model for identifying cancer. The model is being trained on billions of images and can identify common and rare cancers. Paige has been digitizing the pathologists’ workflow to improve accuracy and efficiency within the specialty. The AI model will help solve the storage problem for health systems and help pathologists work through cases and arrive at a diagnosis more quickly.

The IRS uses AI as intended – take complex problems and make your existing teams more capable by leveraging the tech.    Love it.

Healthcare, too – take specialized information and make it available more broadly, particularly in helping primary care physicians get to diagnosis more quickly. 

I’m trying to pull far less hype-driven examples in my coverage to frame the conversation.   These are examples where you can tie quickly to business and direct human outcomes.    We like outcome-driven solutions.