This week marked important progress in how agentic AI is transforming scientific research and discovery. Agentic AI represents a new type of artificial intelligence that can understand goals, make plans, and take actions to help people accomplish complex tasks.

The biggest announcement came from CAS, an American company that works with the American Chemical Society. On October 6th, they revealed new AI agents designed specifically for scientists. These intelligent helpers will be added to their research tools called CAS BioFinder and CAS SciFinder. What makes these AI agents special is that they were trained on real scientific information, not just general knowledge from the internet.

Scientists can now ask these AI agents questions using normal, everyday language, just like talking to a colleague. The AI understands what they need and searches through huge collections of scientific data to find answers. Even better, the AI can respond in different languages, which helps researchers all around the world. When the AI gives an answer, it shows where the information came from so scientists can check if it's correct.

One real example shows how powerful this can be. Researchers studying ovarian cancer used the conversational AI agent to guide their work and discover new ideas. While they could have found this information using older methods, the AI agent made the process much faster and easier. The AI organized the information like a report, making it simple to understand.

Manuel Guzman, who leads CAS, explained that these AI agents save scientists valuable time. Instead of spending hours searching for information, scientists can focus on the creative and important parts of their work. The AI handles the tedious task of looking through mountains of data and finding the most relevant facts.

Another major development happened on October 8th when Thunder, Salesforce, and the University of Southern California joined forces. They launched the USC Agentic AI Research Lab, which is the first of its kind. This lab will be a place where students, business leaders, and researchers work together to understand how agentic AI will change different industries.

The lab will study four important areas: healthcare, how companies use AI, high-tech industries, and government programs. Students will get hands-on experience working on real projects with big companies. They'll also have opportunities for internships and jobs, and they'll meet executives who are shaping the future of AI.

These advances show a clear trend in scientific research. Agentic AI is different from older AI tools because it doesn't just wait for commands. Instead, these AI agents can understand what needs to be done, break big tasks into smaller steps, and work through them systematically. They can also learn from experience and get better over time.

For the scientific community, this means faster discoveries and deeper insights. The AI agents can spot connections between different pieces of information that humans might miss because there's simply too much data to review. They use something called a knowledge graph, which is like a map showing how different scientific concepts relate to each other.

Importantly, these AI systems are designed with scientific standards in mind. CAS built their agents to be transparent, meaning scientists can see how the AI reached its conclusions. There are also special checking systems that flag potential mistakes. Scientists can save their conversations with the AI and review them later, which is important for keeping good research records.

The introduction of agentic AI in scientific research doesn't replace human scientists. Instead, it enhances their expertise and gives them powerful new tools. Scientists still make the important decisions and use their knowledge and creativity. The AI simply handles the time-consuming work of organizing and presenting information.

Looking ahead, CAS plans to expand these capabilities even further. Future versions will help with predictive modeling, which means using data to forecast what might happen in experiments or in nature. This could help researchers design better studies and make more confident decisions about their work.

For students and future scientists, the new research lab at USC provides a glimpse of what's coming. As agentic AI becomes more common in research, young people entering science will need to understand how to work with these intelligent systems. The lab ensures that the next generation of researchers will be prepared for this AI-enabled future.

This weekly update shows that agentic AI is moving from theory to practice in scientific research. What once seemed like science fiction is now becoming an everyday tool that helps researchers around the world make discoveries faster and more efficiently.