×
Chinese AI Pioneer Advocates Responsible Development Amid Geopolitical Tensions
Written by
Published on
Join our daily newsletter for breaking news, product launches and deals, research breakdowns, and other industry-leading AI coverage
Join Now

Here is a summary of the key points from the interview with Zhang Hongjiang, founder of the Beijing Academy of Artificial Intelligence:

China’s advantages and challenges in AI development: Despite recent US chip export controls hampering computing power, Zhang sees China’s key AI advantages as its large pool of talented entrepreneurs, vast market with diverse AI application scenarios, and strong research institutions. However, he notes the field is over-competitive with too many similar startups.

Advocating for responsible AI development: As one of China’s leading voices on AI safety, Zhang has worked to raise awareness among researchers, industry and government about the need to mitigate existential risks from advanced AI systems:

  • He argues AI systems should never be able to deceive humans, have persuasive powers exceeding humans, produce weapons of mass destruction, or recursively self-improve in an uncontrollable manner.
  • Zhang stresses the importance of international collaboration on technical solutions and governance frameworks to ensure AI remains under human control as capabilities rapidly advance.

The transformative potential of large language models: Zhang is excited by how foundation models and multi-modal AI are enabling robots to understand complex commands and reason in sophisticated ways, believing this will be a major breakthrough for the robotics field where China has advantages:

  • He cites impressive demos like robots placing objects on photos of specific people or fetching water when someone says they are thirsty, showcasing nuanced language understanding and reasoning.
  • China’s manufacturing expertise and large industrial datasets could give it an edge in developing AI-powered robotics, Zhang predicts.

Debating open source vs closed models: While some Chinese tech leaders argue keeping models closed is necessary for commercialization, Zhang believes open source and proprietary approaches can co-exist, drawing parallels to historic dynamics between Linux and Windows or Android and iOS.

Looking ahead: No longer leading BAAI day-to-day, Zhang is spending more time overseas engaging governments and the global AI community. He remains committed to bringing researchers together across borders to responsibly advance the technology. But he worries geopolitical tensions are inhibiting the scientific collaboration required to proactively address AI risks.

Zhang Hongjiang, founder of BAAI: ‘AI systems should never be able to deceive humans’

Recent News

Salesforce launches testing center for AI agents

As AI agents proliferate across businesses, companies seek robust testing environments to validate autonomous systems before deployment in mission-critical operations.

Google’s Anthropic deal faces Justice Department scrutiny

U.S. regulators seek to restrict Google's ability to invest in AI startups, marking the first major government intervention in big tech's artificial intelligence deals.

AI will drive major scientific advances, NVIDIA CEO tells SC24

NVIDIA's latest computing tools reduce scientific simulation times from weeks to minutes, making advanced research more accessible to labs and companies.