Microsoft has introduced its Majorana 2 quantum chip, achieving a significant leap in qubit reliability. The chip's development serves as a demonstration of the company's Discovery agentic AI platform, now available for enterprise and research use.
Google has introduced Empirical Research Assistance (ERA), an AI tool aimed at accelerating expert-level scientific research by automating code generation and experimentation. Powered by Gemini technology, ERA has demonstrated strong performance on complex scientific challenges across fields such as genomics and neuroscience. The tool is available for select users through Google Labs’ trusted tester program.
The growing influence of AI in scientific research is upending traditional data storage models, shifting priorities from archival preservation to continuous, high-speed accessibility. Major storage vendors are adapting their offerings to meet these evolving needs, as scientific organizations are compelled to rethink data infrastructure for AI-driven discovery.
Google has launched Gemini for Science, a set of AI-powered tools designed to accelerate and streamline scientific research, at Google I/O 2026. The suite includes features for hypothesis generation, computational discovery, literature analysis, and access to life science databases. Access to these tools will be rolled out gradually, with enterprise offerings also available via Google Cloud.
Two AI-powered assistants developed by Google and FutureHouse have successfully aided scientists with drug retargeting tasks, according to new studies published in Nature. These systems help researchers analyze extensive biological data and generate hypotheses, streamlining the process of identifying potential new uses for existing drugs. Both platforms highlight the growing role of artificial intelligence in accelerating scientific discovery.
Researchers have developed an artificial intelligence tool that allows chemists to invent novel molecules simply by describing desired characteristics in natural language. The system leverages advances in large language models and generative neural networks to bridge the gap between text prompts and molecular structures.
A recent study has shown that quantum artificial intelligence systems can significantly improve predictions of chaotic systems, which have traditionally been unpredictable. The research highlights the role of advanced neural network models in addressing challenges found in areas such as weather forecasting and financial modeling.
A new system, SYNAPS-I, has been launched at U.S. Department of Energy (DOE) labs to integrate artificial intelligence into live experimental workflows. The platform allows real-time data analysis, enabling immediate adjustments during scientific experiments and marking a shift from traditional post-experiment analysis. This development could accelerate discovery and increase the efficiency of advanced research facilities.
MIT has released a report outlining the opportunities and challenges at the intersection of artificial intelligence and the mathematical and physical sciences. The white paper, published following a major workshop, stresses the need for coordinated investment, interdisciplinary training, and community-building to advance both disciplines. The institute is aligning its own research, training, and hiring efforts to realise these aims.
Researchers have combined artificial intelligence and x-ray scanning technology to rapidly generate detailed 3D models of thousands of ant specimens. This approach allows for unprecedented visualization and digital preservation of biological samples, marking a significant advancement in scientific research tools.
Stay informed with curated insights delivered weekly to your inbox.