The upcoming AI showdown between US and China seems to be an inevitability at this point, but apart from jingoism, there might be practical reasons why the US would want to win.
Former US Secretary of State Condoleezza Rice has said that it’s better that a democratic open society wins the AI race, because it’ll have adequate safeguards in case AI goes rogue and something goes wrong. She said that a closed society like China would likely hide these negative effects, only for them to snowball into something bigger later.
“I am one who believes that if somebody’s going to win the race on these frontier technologies, it had better be a democracy,” Rice said on a podcast. “Because if something goes wrong in AI — and it’s quite possible that something will, it’s probable that something will, maybe it’s even predictable that something will — we will have investigative reporting,” he added.
“(We’ll have podcasts talking about it). We’ll have Congressional hearings. The Chinese, instead, will do what they did with Covid. They’ll hide it. They’ll lie about it. And so if an open society that develops these foundational technologies, these transformational technologies, I believe is simply safer for humankind,” he said,
It’s a persuasive argument. An open society like the US with broad provisions for freedom of speech will likely have several outlets for people to register their experiences with AI — they’ll be able to post freely on the internet, podcasts and the press will pick it up, and opposition politicians will be on the lookout to hold the government accountable. In contrast, China has few such outlets for people to air their grievances — the internet is tightly controlled, there’s no free press, and there is no opposition party in the single party state. As such, initial instances of misalignment of models might end up going reported, and lead to much bigger problems later on. China is making rapid strides in AI, with DeepSeek, Moonshot AI and a host of other companies, but its governance structure might not necessarily be ideal for the safe deployment of the technology at scale.