Eli Lilly Inaugurates LillyPod

Eli Lilly has inaugurated LillyPod, the pharmaceutical industry’s most powerful AI supercomputer, built on an NVIDIA DGX SuperPOD with 1,016 Blackwell Ultra GPUs delivering over 9,000 petaflops of performance.

Enterprise Readiness for Agentic AI

A Cisco report surveying 650 executives across six countries found that 80% of business leaders believe their company’s survival will depend on agentic AI by 2027. Executives in the survey predict that 55% of their workforce will be collaborating with AI agents within 24 months.

Among early adopters in the Cisco survey who have invested strategically, 43% report meaningful ROI and 39% anticipate returns within a year.

Strategic Partnerships Accelerating Deployment

Snowflake and OpenAI entered into a landmark $200 million strategic partnership aimed at accelerating the deployment of agentic AI for corporate enterprises.

Medical and Scientific Breakthroughs

UCSF Generative AI Matches Expert Teams

A study published in Cell Reports Medicine by UCSF researchers found that generative AI could handle complex medical datasets on vaginal microbiome data linked to preterm birth risk as well as or better than human expert teams that spent months building prediction models.

Physics-Informed AI Ensures Plausible Outputs

Researchers at the University of Hawaiʻi at Mānoa published a new physics-informed machine learning algorithm in AIP Advances that ensures AI outputs remain physically plausible, even when data is sparse, by embedding the laws of physics directly into the model.

Weill Cornell Launches Major AI Integration Program

Weill Cornell Medicine announced the launch of its AI to Advance Medicine (AIM) program, a major initiative to integrate AI into clinical care and biomedical research focused on tools for precision medicine.

Advancing AI Capabilities

Alibaba Unveils Qwen 3.5

Alibaba unveiled Qwen 3.5, a major upgrade to its flagship model designed for agentic, multimodal tasks, including the ability to analyze videos up to two hours long.

Neuromorphic Computing Solves Complex Physics

Researchers demonstrated that neuromorphic computers, processors modeled after the human brain, can now solve the complex equations behind physics simulations, a capability once thought exclusive to energy-hungry supercomputers.

NASA Mars Rover Driven by AI

NASA’s Perseverance rover completed the first Mars drives ever planned by artificial intelligence, using Anthropic’s Claude vision-language models to analyze orbital imagery and terrain data and autonomously generate safe waypoints.

AI Applications at UC San Diego

Climate Modeling Accelerated 25-Fold

UC San Diego and the Allen Institute for AI developed a new model, Spherical DYffusion, that projects 100 years of climate patterns in just 25 hours by combining generative AI techniques with physics-based data, delivering results 25 times faster than current methods.

Gene Discovery in Alzheimer’s Research

UC San Diego bioengineers used AI to model 3D protein structures and discovered that the gene PHGDH has a hidden role disrupting how brain cells switch genes on and off, a disturbance that can fuel Alzheimer’s disease.

Non-Invasive Cardiac Monitoring

UC San Diego researchers recorded signals from outside heart muscle cells and used AI to reconstruct what is happening inside these cells, monitoring heart activity with remarkable accuracy without invasive methods.

Reducing Radiation Exposure in Cancer Treatment

UC San Diego researchers developed advanced deep-learning techniques that reduced errors in radiation doses to critical organs such as the heart and lungs during breast cancer radiotherapy.

Skin-Mounted Ultrasound Patch Controls Robotic Limbs

UC San Diego engineers created a flexible wireless skin-mounted ultrasound patch that monitors muscle activity in real time to control a robotic arm, with an AI algorithm mapping signals to muscle distributions.

Emergency Response and Public Safety

ALERTCalifornia Harnesses AI for Wildfire Management

ALERTCalifornia operates a network of more than 1,200 natural-hazard monitoring cameras spanning remote mountaintops to wildland-urban interfaces and utilizes AI to help emergency managers spot smoke, monitor fires and plan evacuations.

Gender Bias in AI Recruitment Tools

A comprehensive study conducted in Belgium revealed that gender bias in AI-assisted recruitment tools is far more pervasive than previously thought, with AI models using proxy variables such as hobbies, language patterns, or career gaps to penalize female candidates.

Source: Crescendo AI