Protesters are set to take to the streets this weekend to advocate against the harmful applications of artificial intelligence (AI) and to call on the Australian government to pause advancements in what is being labeled as potentially “the most dangerous technology ever created.” The global protest movement known as PauseAI is mobilizing in cities such as Melbourne in anticipation of next week’s Artificial Intelligence Action Summit in Paris. Demonstrators argue that the summit fails to prioritize AI safety.
Both China and the United States are rapidly advancing their AI development efforts. Millions of Australians are utilizing the Chinese AI application DeepSeek, while the US is investing $500 billion to boost its own AI initiatives through a collaboration called Stargate. DeepSeek, a Chinese-owned open-source AI platform, was developed with significantly less investment, time, and infrastructure compared to its American tech counterparts.
Joep Meindertsma, the founder of PauseAI, initiated this political movement in 2023, expressing concerns that companies like OpenAI and DeepSeek are not implementing sufficient safety measures for their AI models before releasing them. The protesters are advocating for the establishment of an international AI Pause treaty, which would suspend the training of AI systems more advanced than GPT-4 until they can be developed safely and democratically.
“It’s no longer a secret that AI could be the most dangerous technology ever created,” Meindertsma stated. “We need our leaders to act on the slim chance that things could go very wrong, very soon. Our psychology makes it challenging to recognize and respond to these threats. Abstract dangers like this one don’t seem to alarm people as they should.”
The protest highlights increasing worries that AI development is outpacing society’s ability to manage it, especially as experts still grapple with understanding the complexities of AI systems like ChatGPT. Recently, Google discreetly updated its AI ethics, no longer excluding military or surveillance applications.
Meindertsma noted that the three most prominent AI researchers—Geoffrey Hinton, Yoshua Bengio, and Ilya Sutskever—have all publicly acknowledged that the technology could potentially lead to human extinction. In December, the Minister for Industry and Science, Ed Husic, announced a National AI Capability Plan, which is expected to be completed by the end of 2025.
Meindertsma emphasized the necessity of global action at the summit, rather than relying on individual nations like Australia to implement safety measures. He warned that without meaningful international regulations, the organizers risk ignoring the critical need for safety.
Next week’s summit follows previous gatherings in Bletchley, England, and Seoul, Korea, which resulted in the Bletchley Declaration and the Seoul Declaration. Notably, Federal Industry and Science Minister Ed Husic will not be attending the Paris summit.