Artificial intelligence is evolving at an unprecedented pace, raising urgent questions about its impact on society. One of the latest entrants to the AI arms race is DeepSeek AI, a powerful open-source Chinese model capable of advanced reasoning, content generation, and data analysis.
As Australia grapples with how to regulate AI responsibly, the question has arisen about whether we should restrict the use of DeepSeek.
Many have already called for a TikTok-style ban, where public officials won’t be able to download and use the app.
As the AFR warned, ‘China could gain a competitive edge in the global tech arms race against the USA that could have far-reaching global implications about who writes the rules of AI.’ But even if China does win, does that automatically mean we’ll see ‘Chinese political values and norms – which are often antithetical to liberal democracies – shaping the AI revolution’?
Those arguing for restrictions – especially when it comes to use in the public sector – cite risks to Australian national security, government operations, and private-sector intellectual property.
Further, there has been a growing concern worldwide that DeepSeek could be exploited to spread misinformation, manipulate public opinion, or interfere in democratic processes. That said, it is difficult to argue that the risks presented by DeepSeek are any different than ChatGPT or Claude in the hands of a malicious actor.
Certainly, those trying it over the past few days have found that in DeepSeek’s world, there is no record of the Tiananmen Square massacre or other atrocities committed by the CCP. However, it’s long been common knowledge that certain topics are similarly – although not as completely – suppressed on the China-owned social channel TikTok. So while the new tool did not come with a caveat emptor label, its bias should not have come as a surprise.
The truth is that every digital algorithm will have some level of bias, despite claims to the contrary. Australia should take pause before be taking sides in the China vs US race for AI supremacy. Currently, DeepSeek’s only confirmed crime appears to be outperforming many of its American competitors.
Placing blanket restrictions on AI use risks setting a precedent for excessive government control over emerging technologies. Many argue that instead of outright bans, Australia could focus on establishing ethical AI guidelines and improving digital literacy to ensure responsible usage.
By doing so, Australia can embrace AI’s benefits while safeguarding its institutions and citizens against potential harm.
As a country rooted in liberal democratic values, Australia should avoid resorting to authoritarian tactics, even in the defense of those same liberal democratic values.
In the end, competition – not allegiance to any one global power – offers Australia its best defense in the rapidly evolving tech landscape. Let’s avoid knee-jerk reactions and make decisions based on a sober assessment of risks and benefits, free from ideological bias.
Emilie Dye is a Marketing and Research Analyst for the Centre for Independent Studies.