A Canadian artificial intelligence expert says society should ensure that AI is “properly governed” amid a push in the United States to place guardrails on the rapidly advancing technology. This week, key players from major AI and technology companies wrote an open letter urging the U.S. government to support a global effort to develop tools that can “deliberately pace the frontier of automated AI development.” The signees wrote that “industry, government, and society at large may need the option to buy time to address emerging risks, develop security measures, and strengthen oversight.” The letter was signed by more than 1,000 staff, including the chief scientist of OpenAI, one of the ChatGPT developer’s original co-founder, and vice-presidents at Meta and Google, CNN reported. The petition was created in the wake of an “unprecedented” breach by a rogue OpenAI autonomous agent. “There should be caution and hesitancy to make sure this technology is properly governed, because things like the hack that OpenAI system did can happen,” Mark Daley, chief AI officer at Western University in London, Ont., said in an interview with CTV Your Morning on Thursday. However, Daley suggests AI can also provide massive benefits, including in the medical field. “Those very same systems could help us discover cures to rare diseases that we otherwise wouldn’t have the bandwidth to research,” he said, adding there’s potential for “so much possible public good” to come from AI. “That’s what excites me.” U.S. lawmakers propose ‘AI kill switch’ bill Due to security risks from the rising power of AI, U.S. lawmakers have recently proposed an “AI Kill Switch Act,” enabling federal officials to require AI companies to shut down models that pose a danger to humans and the economy. A bipartisan group in Washington also proposed a bill that would require independent security audits for powerful AI models. But these efforts may face pushback. Daley said the petitioners may find it challenging to get the U.S. government’s support, because of the “geostrategic race” between the U.S. and China. “If the U.S. slows down, China could move ahead, and that’s obviously not in the interest of the U.S.,” he said. Daley said the world needs a global mechanism that requires participation from countries, similar to nuclear weapons and arms control verification. This mechanism would “verify what level of capability everyone has, and that’s the lynchpin ensuring a safe future,” he said. With files from CNN and Reuters Watch the video above for the full interview
