The rise of generative AI is reshaping university classrooms, creating both excitement and uncertainty. Lecturers are being asked not only to introduce new technology but also to rethink how students engage with material, develop critical thinking, and complete assignments. For many, the sheer pace of AI development makes it difficult to know where to start or which approaches are both practical and academically sound.
While AI tools promise efficiency and creativity, their application in higher education requires careful consideration. Educators need resources that help them integrate AI responsibly, maintain scholarly standards, and provide students with accurate, evidence-based information. Without this support, attempts to incorporate AI can feel fragmented, time-consuming, or misaligned with learning objectives.
The AI research agent scienceOS is is designed to meet these needs. By giving lecturers access to an extensive, research-backed knowledge base, offering collaborative tools, and generating outputs with full references, it provides a practical, grounded way to explore AI in the classroom. Rather than replacing academic judgment, scienceOS supports it, allowing educators to experiment with AI-enhanced teaching while maintaining transparency, rigor, and control over the learning experience.
Using AI to plan lessons
Even unspecialized AI chatbots are improving at distinguishing scientific facts from common misconceptions Richter, 2025, offering educators promising tools to design lessons. Yet this promise relies on critical use: AI systems may still echo incorrect statements rather than correct them, so it is essential for educators to approach them carefully and to formulate clear prompts.
By sharing this critical awareness with their students, educators can promote responsible and ethically informed use of AI: such as verifying sources, citing AI contributions properly , and checking information for accuracy.
When integrated into classroom activities, collaborative AI tools can enhance learning by fostering human-AI cooperation. Groups of students may join forces with the AI on problem-solving tasks or research projects. This makes lessons more interactive and helps build skills like teamwork, critical thinking, and evidence-based reasoning.
Get scientific answers. A short video showing how the AI research agent provides scientific answers grounded in published research.
In this context, scienceOS demonstrates how AI can support rigorous and collaborative science education. Its responses are grounded in verified scientific literature, reducing the risk of misconceptions, and its shared workspaces facilitate collaborative engagement. By combining transparency, evidence-based reasoning, and collaboration, scienceOS aligns closely with pedagogical and ethical considerations.
To employ scienceOS in designing lessons based on reliable, evidence-based information, follow this brief workflow:
- Create a new project: Use the “New project” button on the app homepage or click the plus button in the sidebar to create a new project. Give the project a clear name.
- Add sources to the project: During the project setup, you can add individual documents, one or more collections, or your entire library to the project sources. If you do not have any documents in your library, ignore this and simply make sure to have “230 million research papers” enabled. (You can still modify what the AI can access after creating the project.)
- Invite and manage project members (optional): You may safely skip this step for preparing lessons. (You can still invite your colleagues or students to the project at a later stage.)
- Ask questions: Start a chat about your area of interest. Add context to your prompt to make sure the AI can understand your goal and intentions. For example, you may ask for a topical outline or an answer that includes a relatable analogy. The AI agent will generate answers with references to relevant literature.
You can review these sources, use them to build a personal library, and prepare classes with verified, current materials. This ensures that lessons remain accurate, engaging, and grounded in evidence.
Sharing projects with students
ScienceOS makes it easy to share projects with students in a controlled and educational way. By preparing a shared project, educators can give students access to carefully selected sources and model examples of well-structured prompts. This shared environment helps guide students toward responsible use of AI while still leaving room for exploration and collaborative learning within clearly defined boundaries.
Projects also allow educators to customize the behavior of the AI agent. For example, the system can be prompted not only to answer student questions but also to provide directions for further exploration, suggest analogies, and guide students toward a deeper understanding of the subject. This ensures that AI support aligns with the learning goals you set and helps students develop the skills to use AI thoughtfully in their academic work.
- Upload papers
Upload the most relevant or easy-to-digest literature on your topic to your scienceOS library. (Skip if you already have sources saved to your library.)
- Create a new project
Create a new project from the app homepage or click the plus button in the sidebar to create a new project. Give it a compelling name.
- Add sources to the project
Add the previously uploaded literature to the project sources.
- Create model prompts (optional)
Ask some questions to the project to construct examples for good prompting, generate a summary of the sources in the project, or highlight a specific feature such as diagrams and tables. Click the “star” button to save the prompts and related answers to the project feed. All project members will be able to see and interact with these saved answers.
- Invite students
Click “Share” in the top right corner of the project homepage. Copy the invitation link and send it to your students. (You may also turn the invitation link into a QR code and add it to your slides.)
Customizing the AI for education. A brief video showing how to customize the behavior of the AI agent within a project to align with the learning goal of a lecture series, and how a saved answer can be used as a starting point for further literature exploration.
By adding saved answers to a shared project before inviting students, educators can provide examples for careful prompting, understanding AI responses, and inspiring creative thinking. Students can use these saved responses as a starting point for their own work, and the shared project feed encourages active discovery and collaboration.
This approach can even facilitate a journal-club–style reading group. By sharing a curated set of papers, students can explore topics using scienceOS and save their AI-generated responses to the project feed, allowing everyone in the group – students and instructors – to review, compare, and discuss different perspectives collaboratively. In this way, individual reading becomes an interactive, discussion-driven learning experience.
Collaboration with colleagues
ScienceOS also supports professional collaboration among educators. For example, to collectively develop teaching materials, exchange ideas, or implement different instructions and customized behaviors for various groups or subjects. This flexibility allows teams of educators to adapt AI support to their specific needs, fostering a more dynamic and cooperative teaching environment.
All generated answers include cited sources, which can be reviewed and added to shared or personal libraries. Educators can also add custom AI actions to follow new research and update their teaching strategies.
Although scienceOS is primarily designed for scientists, it is also an excellent tool for education. It promotes transparency, evidence-based thinking, and collaboration. By exploring and verifying information together, teachers and students can build a classroom culture that values curiosity, accuracy, and critical thinking.
