AI Tool Developed by University at Buffalo Researchers Outperforms Doctors and Other AI Tools on USMLE Exams

Researchers at the University at Buffalo developed Semantic Clinical Artificial Intelligence (SCAI), an AI tool that scored 95.2% on Step 3 of the USMLE, outperforming other AIs like GPT4 Omni (90.5%) and most doctors. Using semantic reasoning with a knowledge base of 13 million medical facts, SCAI can engage in conversations to provide accurate answers. Designed to augment rather than replace physicians, it aims to enhance patient safety, improve care access, and democratize specialty care.

SCAI Outperforms Other AI Models in Medical Question Answering

SCAI leverages semantic reasoning to process complex medical information with precision. It constructs semantic networks using 13 million medical facts and triples, such as “Penicillin treats pneumococcal pneumonia,” enabling more effective reasoning in clinical contexts. This structured approach enhances accuracy and reliability compared to traditional generative AI tools that rely on pattern recognition.

The integration of knowledge graphs further improves SCAI’s capabilities by mapping relationships between medical concepts. These graphs facilitate a deeper understanding and application of evidence-based medicine, ensuring responses are grounded in current data and reducing errors. Retrieval-augmented generation ensures up-to-date information is used to refine answers, increasing trustworthiness.

SCAI’s semantic reasoning supports handling nuanced clinical questions by analyzing context and relationships within medical knowledge. This capability is particularly valuable in specialties requiring complex decision-making, as it helps bridge gaps between general and specialized care. By providing access to specialized medical information, SCAI empowers primary care providers and patients, democratizing specialty care and enhancing healthcare accessibility.

Lead researcher Peter Elkin emphasizes that AI tools like SCAI are designed to augment, not replace, physicians. This perspective addresses concerns about AI replacing human roles, highlighting its role as a supportive tool in advancing medical practice through technology.

SCAI leverages semantic reasoning to process complex medical information with precision. It constructs semantic networks using 13 million medical facts and triples, such as “Penicillin treats pneumococcal pneumonia,” enabling more effective reasoning in clinical contexts. This structured methodology enhances its ability to reason accurately in clinical scenarios compared to traditional generative AI tools that rely on pattern recognition.

The integration of knowledge graphs further improves SCAI’s capabilities by mapping relationships between medical concepts. These graphs facilitate a deeper understanding and application of evidence-based medicine, ensuring responses are grounded in current data and reducing errors. Retrieval-augmented generation ensures up-to-date information is used to refine answers, increasing trustworthiness.

SCAI’s semantic reasoning supports handling nuanced clinical questions by analyzing context and relationships within medical knowledge. This capability is particularly valuable in specialties requiring complex decision-making, as it helps bridge gaps between general and specialized care. By providing access to specialized medical information, SCAI empowers primary care providers and patients, democratizing specialty care and enhancing healthcare accessibility.

Lead researcher Peter Elkin emphasizes that AI tools like SCAI are designed to augment, not replace, physicians. This perspective addresses concerns about AI replacing human roles, highlighting its role as a supportive tool in advancing medical practice through technology.

More information
External Link: Click Here For More

Quantum News

Quantum News

As the Official Quantum Dog (or hound) by role is to dig out the latest nuggets of quantum goodness. There is so much happening right now in the field of technology, whether AI or the march of robots. But Quantum occupies a special space. Quite literally a special space. A Hilbert space infact, haha! Here I try to provide some of the news that might be considered breaking news in the Quantum Computing space.

Latest Posts by Quantum News:

Toyota & ORCA Achieve 80% Compute Time Reduction Using Quantum Reservoir Computing

Toyota & ORCA Achieve 80% Compute Time Reduction Using Quantum Reservoir Computing

January 14, 2026
GlobalFoundries Acquires Synopsys’ Processor IP to Accelerate Physical AI

GlobalFoundries Acquires Synopsys’ Processor IP to Accelerate Physical AI

January 14, 2026
Fujitsu & Toyota Systems Accelerate Automotive Design 20x with Quantum-Inspired AI

Fujitsu & Toyota Systems Accelerate Automotive Design 20x with Quantum-Inspired AI

January 14, 2026