GitLab

One Australian business has discouraged personnel from using the technology, mariskamast.net others are rushing for advice on its cybersecurity implications - while federal government ministers are urging care.

But others have invited DeepSeek's arrival, calling for Australia to follow China's lead in establishing powerful yet less energy-intensive AI innovation.

In the days considering that the Chinese business introduced its R1 artificial intelligence model and publicly released its chatbot and app, it has actually upended the AI market.

- Sign up for Guardian Australia's breaking news email

Several international industry leaders saw their market price drop after the launch, as DeepSeek showed AI could be developed using a portion of the expense and processing required to train models such as ChatGPT or Meta's Llama.

Its arrival might signify a new market shift, but for government and business, the result is unclear. Whereas ChatGPT's 2022 arrival caught governments and services by surprise as staff began to experiment with the new AI technology, at least for the arrival of Deepseek, rocksoff.org some had a playbook.

Business as usual

A spokesperson for Telstra stated the company had "a strenuous procedure to evaluate all AI tools, capabilities, and use cases in our company", consisting of a list of authorized generative AI tools, and guidelines on how to use them.

For now at Telstra, DeepSeek is not approved and its usage is not motivated (although it's not officially blocked).

"Our preferred partner is MS Copilot, and we're rolling out 21,000 Copilot for Microsoft 365 licences to our workers."

Other business sought immediate advice on whether DeepSeek must be embraced.

Major Australian cybersecurity firm CyberCX's executive director of cyber intelligence, Katherine Mansted, said had actually already approached the company for advice on whether the technology was safe.

"That's no surprise, due to the fact that it appears the entire world has remained in a bit of a DeepSeek frenzy - both the economically and market inclined and those with the security lens," Mansted said.

DeepSeek and federal government

CyberCX today took the unusual action of rapidly issuing advice suggesting organisations, consisting of federal government departments and those keeping delicate details, strongly consider limiting access to DeepSeek on work gadgets.

"We understand that there is no proactive policy here from federal government ... We've been down this road previously," Mansted said. "We've had disputes about TikTok, about Chinese monitoring video cameras, about Huawei in the telco network, and we constantly act after the fact, not before the fact ... Here, especially since the hazards are around compromise of sensitive info, in regards to any info that you put into this AI assistant: it's going directly to China.

"We thought we needed to act quicker this time."

Under federal AI policy executed in September 2024, companies have up until completion of February 2025 to publish openness documents about their usage of AI.

But understanding who makes choices on the particular use of DeepSeek in the federal government has shown tricky. The attorney general of the United States's department, which made the choice to ban TikTok utilize on federal government gadgets, referred questions to the Digital Transformation Agency, which in turn referred enquires to the Department of Home Affairs.

Home Affairs was asked on Thursday for its official policy and did not offer an action by the time of publication.

Familiar debates ...

A few of the reaction in Australia to DeepSeek is by now familiar. There have actually been calls to ban the technology, in the middle of concern over how the Chinese federal government might access user information - an echo of the days Huawei was banned from the NBN and 5G rollouts in Australia, and more just recently, of the debate over banning TikTok.

The Australian Strategic Policy Institute, a strong critic of the China federal government, stated today that Australia "can not continue the present technique of reacting to each new tech advancement". It required a tech technique covering AI that included investing in sovereign AI abilities.

The industry minister, Ed Husic, said on Tuesday it was prematurely to decide on whether DeepSeek was a security threat.

Register to Breaking News Australia

Get the most important news as it breaks

"If there is anything that provides a risk in the national interest, we will constantly keep an open mind and view what takes place. I think it's prematurely to leap to conclusions on that," he said. "But, again, if we have to act, then responsible federal governments do."

He worried that Australia is "in the last stages" of preparing its action and would establish its own regulative settings.

"The US is flagging their approach. The EU has theirs. Canada likewise will have a various approach. And our regional partners also are looking at this," he said.