Doctoral Student Experimenting Lab: Explore the responsible use of AI in generating scientific texts, images, audio and code

Abstract

Explore Generative AI for Responsible Scientific Work. Through practical exploration guided by our expert lecturers and tutors, attendees will discover how generative AI can create texts, images, and audio—each with applications spanning diverse scientific contexts, from extracting data and data analysis to scientific illustrations and scholarly publications. This course goes beyond mere technical proficiency; it emphasizes responsible and ethical usage of Large Language Models (LLMs).

Lernziele

The course learning objectives for the labs are designed to ensure participants gain practical skills and understanding in the following areas: First, they will receive hands-on training to use generative AI responsibly, efficiently, and professionally in daily scientific work. Second, they will develop competencies to navigate the limitations, potential pitfalls, and benefits of AI-based tools, with a focus on maintaining research integrity and ethics. Additionally, participants will gain experiential knowledge of generative AI by actively experimenting with AI technologies.

Inhalt

This 4-day course for PhD students allows for experimenting with generative AI, enabling the creation of texts, images, and code for scientific contexts like presentations and publications. Utilizing a problem-based learning approach, participants engage with critical issues related to AI-based tools, including the use of research data, private data, and the generation of diverse outputs such as images, text, and code. Invited experts will conduct hands-on workshops focusing on generating scientific illustrations and customizing AI models, such as GPTs and bots. The course emphasizes managing biases in AI outputs and upholding ethical standards, ensuring that the application of AI aligns with research integrity and the values of the scientific community. Students will explore technologies designed to protect personal and copyrighted data, and gain insights into ensuring reproducibility and transparency in AI-generated outputs. The course addresses key concerns while using AI tools.

Ort

• ETH Zentrum
  Alumni Pavillon (MM C 78.1)

Zielgruppe

Sprache

Kommentar

Do you have specific questions related to the use of generative AI for scientific purposes? Are you currently customizing GPTs to facilitate your own workflows? We’d love to hear from you! Contact the organizers at psc_phdprogram@ethz.ch and get involved in creating a course that will benefit our community.