AI in Healthcare

Today's Stories

1. 1

   Jacobi–IDE, a development tool for Abaqus subroutines, now runs analytical physics tests automatically and can send failed diagnostics to Claude for AI-powered debugging.

   Jacobi–IDE compiles and runs a suite of 12 physics tests on Abaqus subroutines (specialist engineering simulation code), flagging errors in material properties, yield behavior, and numerical stability. When tests fail, the tool sends diagnostic data to Claude, an AI assistant, which suggests the likely cause—such as a volumetric component in plastic flow that needs correction. Writing and validating Abaqus subroutines is error-prone and time-consuming; engineers typically must debug by hand. Automated analytical tests catch physical inconsistencies early, and AI diagnosis narrows the search space for fixes, reducing iteration cycles and the expertise barrier for less experienced developers.

   The tool is demonstrated passing 10 of 12 tests on an elastic–plastic material model, with Claude correctly identifying a suspected issue in the failed cases. Availability, pricing, and compatibility with different Abaqus versions are not stated in the announcement.

2. 2

   Eli Lilly and Twist Bioscience are taking different industrial approaches to AI-powered drug discovery, signaling how the sector is beginning to operationalize AI beyond research.

   Eli Lilly and Twist Bioscience have adopted distinct strategies to industrialize AI in drug development. Eli Lilly is building internal AI capabilities and infrastructure, while Twist Bioscience is positioning itself as an enabling partner to other organizations in the space. Drug discovery has historically relied on years of lab work and trial-and-error. These two companies' divergent paths show that AI is starting to reshape how the industry actually manufactures and discovers drugs at scale, moving from theory into production.

   The success of these approaches will determine whether AI in drug discovery becomes a competitive advantage that individual pharma firms build in-house, or a shared infrastructure that partners like Twist Bioscience provide to the industry.

3. 3

   AI drug discovery is shifting focus from tool-makers to real-world users—two healthcare stocks show how companies are turning algorithms into actual medicines.

   Eli Lilly and Twist Bioscience are using AI in drug development in different ways. Lilly partners with OpenAI on drug discovery and has built an AI co-innovation lab with Nvidia to find molecules faster; Twist makes synthetic DNA that drugmakers test, and is now integrated into Amazon's new Bio Discovery platform for AI-driven drug design. The first wave of AI rewarded companies that sold the tools. The next wave appears to reward companies that apply those tools in clinics, labs, and pharmacies to produce new medicines. Both firms show how AI adoption is moving into real-world use rather than staying at the software level.

   Eli Lilly's recently approved weight-loss pill Foundayo is a near-term growth driver because a pill format (replacing injections) could reach a much larger patient population. Twist reported revenue rising 19% to $110.7 million（約180億円） in fiscal Q2 2026 and has posted 13 straight quarters of rising revenue, though it currently operates at a net loss.

4. 4

   Merck and Protillion partner on AI-powered drug discovery with up to $510 million（約820億円） in potential milestone payments, signaling Big Pharma's deepening reliance on AI to accelerate early-stage research.

   Merck and Protillion announced a collaboration in which Protillion will apply its AI platform to help Merck discover and develop new drug candidates. The deal includes up to $510 million（約820億円） in milestone payments, in addition to undisclosed upfront and royalty terms. Pharma companies face lengthy and expensive drug development timelines. By partnering with an AI specialist, Merck is betting that computational approaches can reduce the time and cost of identifying promising compounds — a critical competitive advantage in an industry where speed to market directly impacts returns on research investment.

   The agreement spans multiple therapeutic areas and runs through 2027, meaning results and milestone achievements will unfold over the next several years. Success here could influence how aggressively other major pharmaceutical firms pursue similar AI collaborations.

5. 5

   Pramaana Labs raises $27 million（約43億円） to add mathematical verification to AI systems used in law, drug discovery, and tax preparation, where errors are costly.

   Pramaana Labs announced $27 million（約43億円） in seed funding led by Khosla Ventures, with participation from Accel, BoldCap, Nexus Venture Partners, Premji Invest, and Unbound. The startup will use formal verification tools—drawing on the LEAN programming language used to verify mathematical proofs—to add a deterministic verification layer on top of conventional LLMs, checking the AI's reasoning before it reaches users. Enterprises are struggling to turn AI pilot programs into reliable business tools, especially in high-stakes fields where errors are costly. Pramaana's approach addresses this by formalizing the rules of domains like tax law and drug discovery into executable code, making the AI's reasoning deterministic and verifiable rather than opaque. This addresses a core pain point: current AI systems lack sufficient protection against hallucinations and errors in sensitive verticals.

   Pramaana is building domain-specific verification systems, starting with tax law (working with former IRS commissioner Danny Werfel) and cybersecurity and drug discovery (overseen by professors from IIT Delhi, IIT Madras, and UC Berkeley). The company's model requires codifying the rules of each industry before verification can work, making execution and domain expertise the critical path forward.

6. 6

   Google's medical AI system AMIE now handles long-term disease management, matching primary care doctors on clinical reasoning in a new study.

   Google published research in Nature showing that AMIE, its medical AI system, can now manage ongoing health conditions—not just diagnose them. In a blinded study comparing AMIE with 21 primary care doctors using patient actors, AMIE matched clinicians in overall management reasoning and scored significantly higher in plan preciseness and guideline alignment. The shift from one-time diagnosis to continuous disease management is a substantial step toward AI that could reduce the burden on physicians. If AI handles routine tracking of symptoms, medication adjustments, and clinical guideline updates—tasks that consume significant appointment time—doctors may have more capacity to focus directly on patients rather than administrative work.

   Google is launching a nationwide study to test AMIE in real-world virtual care settings. This real-world validation will show whether the AI's lab performance translates when used by actual patients and clinicians outside a controlled trial environment.