Pharma companies are collaborating to boost the power of artificial intelligence (AI) in drug discovery by allowing access to proprietary structural data to train a large language model. Each of the partners is contributing data from several thousand experimentally determined protein:ligand interactions, creating one of the most diverse datasets and the richest chemistry assembled to date for model training.

AI seems to suggest that a world of problems with health care spending may become more manageable, but Stephen Bittinger, a shareholder in the D.C. office of the law firm of Polsinelli PC told BioWorld that all the headaches surrounding validation of these algorithms suggests a need for an independent AI validation institute.

A new generative AI model trained on UK Biobank data can simultaneously forecast the risks and timing of developing over 1,000 different diseases a decade into the future. The developers applied similar algorithmic concepts to those used to develop large language models like ChatGPT and Gemini to build the model, using medical records from 402,799 participants in UK Biobank.

China’s National Medical Products Administration is arguably more active than regulatory agencies in many nations in advancing guidance for AI in medical technology, but Chang-Hong Whitney, President/CEO of Whitney Consulting Ltd., told BioWorld that the premarket review process still carries some unpredictability.

The first phase of the EU’s Data Act is now officially in place, although member state enforcement of the Act might be spotty. This is because several nations, including Germany, have yet to finalize implementing regulations for the Data Act, a problem these nations will have to fix as the secondary and tertiary compliance dates arrive in August 2026 and August 2027.

Companies developing AI-based health care solutions need to ensure they have quality data, a solution that can easily integrate into clinical workflow systems and teams with evolving skills, to attract investment, delegates heard at the LSI Europe ’25 conference in London.

The U.S. FDA cleared Respiree Pte. Ltd.’s 510(k) for its RS-001 cardio-respiratory wearable that measures respiration and offers passive cardio-respiratory monitoring.

The Sept. 3 congressional hearing on AI in health care raised questions about whether the U.S. FDA enjoys the statutory authorities it needs to properly regulate this class of products, but witnesses at the hearing said a lack of trust in AI is a substantial roadblock to adoption.

Patients often wait for a long time to see a mental health specialist to get the help they need, and Greymind AI’s mental health platform, Reboot AI, offers a stop-gap measure to provide structured early intervention with built-in escalation to human clinicians.

Lee Dae-hong founded Aivis Inc. in January 2021 as an AI-powered pathology software company, having led the development of the world’s first commercial under-display fingerprint recognition algorithm that was later incorporated into the Galaxy S10 and Galaxy A series of smartphones of Samsung Electronics Co. Ltd.