#1 — AI Tools for Accelerating Scientific Research

Google has expanded access to AI tools for science:

• Experimental Gemini for Science tools are available via Google Labs for researchers.
• Enterprise-grade Gemini for Science solutions are offered through Google Cloud.

Regarding Co-Scientist:

• The system has not achieved autonomous scientific discovery or replaced human research teams.
• Its current results remain preclinical.
• Google has integrated Co-Scientist into the Gemini app, with a stated goal to reduce research process times from months to hours.

Another Google system, Gemini Deep Think:

• Addresses professional research problems in mathematics, physics, and computer science, operating under expert guidance.
• Achieved a gold-medal standard at the International Mathematics Olympiad.
• Demonstrated comparable results at the International Collegiate Programming Contest (dated 2025).
• Is now applied to open-ended challenges across science, engineering, and enterprise workflows.

Quantum computing timelines show acceleration:

• The Majorana 2 quantum chip moves the anticipated availability of a scalable quantum computer to 2029.
• Microsoft projects a 2029 timeline for a scalable quantum computer, halving prior estimates, attributing this to the Majorana 2 chip and its Discovery capabilities.

FutureHouse, a company backed by Eric Schmidt, released four free AI agents for scientific research:

• Named Crow, Falcon, Owl, and Phoenix, these agents are designed for tasks like analyzing research papers, designing drug compounds, identifying research gaps, and accelerating discovery processes.

Sources

1. New AI Tools for the Future of Science - Google Blog

   Google is introducing Gemini for Science, a collection of AI tools and experiments designed to enhance scientific exploration. These tools, including Hypothesis Generation, Computational Discovery, and Literature Insights, aim to accelerate the scientific method by helping researchers with tasks like synthesizing information, testing hypotheses, and analyzing literature. Access to these experimental tools will gradually open on Google Labs, with enterprise-grade solutions also being deployed through Google Cloud.

2. Google Co-Scientist Reaches Nature: Hypothesis Agents Validated in Lab, Not Yet in Clinic

   Google has launched "Hypothesis Generation," a public-facing tool built on Co-Scientist, its multi-agent AI system for scientific hypothesis generation. The system has been validated in laboratory experiments, with the Nature paper documenting its architecture and results. While significant for generating testable hypotheses in areas like drug repurposing for acute myeloid leukemia and liver fibrosis, Co-Scientist has not achieved autonomous scientific discovery or replaced research teams, and results are preclinical, not clinical.

3. Google announces at I/O the publication of research on its Co ... - Digg

   Google announced at I/O the publication of research on its Co-Scientist multi-agent AI system in Nature, alongside its availability in the Gemini app for science. Co-Scientist aims to accelerate scientific discovery by compressing the research process from months to hours, enabling AI to engage in rigorous scientific reasoning through self-debates. This system, validated with scientists across over 100 institutions, is leading to new insights in areas like antimicrobial resistance, plant immunity, and treatments for various diseases.

4. Gemini Deep Think: Redefining the Future of Scientific Research ...

   Under the direction of expert mathematicians and scientists, Gemini Deep Think is solving professional research problems across mathematics, physics, and computer science. It achieved Gold-medal standard at the International Mathematics Olympiad (IMO) and later, similar results at the International Collegiate Programming Contest in 2025. Subsequently, it has been applied to tackle more complex, open-ended challenges in science, engineering, and enterprise workflows.

5. Majorana 2, made more reliable with Microsoft Discovery agentic AI

   Microsoft has unveiled Majorana 2, a next-generation topological quantum chip that achieves a 1,000-fold improvement in qubit reliability, with a mean qubit lifetime of 20 seconds. This advancement, aided by Microsoft Discovery's agentic AI, accelerates the timeline for a scalable quantum computer to 2029. The company also announced the general availability of Microsoft Discovery, a platform for Frontier R&D that uses AI agent teams to speed up scientific discovery.

6. Microsoft Discovery Reaches GA on Azure, Powering the Agentic AI Behind Majorana 2 Quantum Chip

   Microsoft has announced the general availability of Microsoft Discovery, an Azure-based platform for deploying autonomous AI agent teams in scientific and engineering R&D. The company also unveiled Majorana 2, a next-generation topological quantum chip, whose 1,000-fold reliability improvement was partly achieved using Discovery's agentic AI capabilities. This advancement is expected to halve the timeline for delivering a scalable quantum computer to 2029.

7. Google's Ex CEO Just Released SUPERINTELLIGENT AI AGENTS ...

   FutureHouse, backed by Google's former CEO Eric Schmidt, has launched four free superintelligent AI agents—Crow, Falcon, Owl, and Phoenix—built specifically for scientific research. These AI tools can analyze millions of research papers, design new drug compounds, identify research gaps, and dramatically speed up discovery processes. With transparent reasoning, real-time data integration, and lab testing support, FutureHouse aims to revolutionize science using open, AI-powered systems.

8. Google DeepMind's Co-Scientist Graduates from Research Demo to ...
9. Co-Scientist: A multi-agent AI partner to accelerate research — Google DeepMind