2024-11-26-清华大学-2024年AIGC发展研究报告3.0版_181页_17mb

AIGC Development Research at Tsinghua University

Team Overview

The report is authored by the team of Prof. Shenyang from Tsinghua University's School of Journalism and Communication and School of Computer Science. Led by Prof. Shenyang, a renowned expert in multiple fields including news, AI, and media, the team consists of nearly 40 members and focuses on interdisciplinary research. They emphasize a holistic approach with strong ties to industry and societal contributions. Notable achievements include numerous AIGC implementations and partnerships.

Key Research Areas and Applications

The team explores cross-domain AIGC development, including:

• Media and Communication: Projects like AI-generated news, videos, and content for social platforms, with applications in national exhibitions and micro-videos.
• Healthcare: AI-driven diagnostic tools and collaborative systems for patient care, enhancing precision in clinics and treatments.
• Education: AI-assisted learning models supporting personalized education, such as AIGC-powered writing, programming, and creative tools for students and professionals.
• Creative Arts: Innovations in music (e.g., AI-generated compositions), visual arts (e.g., AI paintings and designs), and literature, with examples like the first AI drama and poems in Chinese.
• Other Domains: Applications in marketing (e.g., RPA-AI integration for efficiency), urban development (e.g., AI city models), and cultural preservation (e.g., AI restoration projects).

Major Innovations and Findings

• Technology Advancements: Development of AIGC tools like Stable Diffusion for images, Runway for videos, and healthcare models, with focus on real-time decision-making and data analysis.
• Ethical Considerations: Addressing challenges such as AI hallucinations, privacy issues (minimizing data misuse), and policy impacts, including the potential for human-AI collaboration and societal risks.
• Global Integration: Partnerships with entities like Google, Apple, and Meta, contributing to AI models that span music, healthcare, and daily consumer apps.

Future Outlook

Future research emphasizes AI-human collaboration, with predictions on AI evolution following phases like sugar-coated AI to advanced AGI. Challenges include managing ethical dilemmas, ensuring equitable AI distribution, and fostering new societal structures. The team advocates for proactive policy in AI governance to maximize benefits while mitigating risks like unemployment and information overload.