One in four companies banning AI ‘for now’ amid privacy concerns: Cisco study

Communications tech giant Cisco has revealed more than 27 per cent of Australian companies are pumping the brakes on AI — for now.

More than a quarter of Australian companies have banned artificial intelligence tools in the workplace, citing privacy concerns — including the potential of corporate secrets being shared with competitors, according to a new study.

Communications technology titan Cisco surveyed more than 2600 security and privacy professionals across 12 countries, which revealed businesses were unplugging the much-hyped technology — for now.

This is despite bigger businesses beginning to reap productivity gains from the much-hyped technology, which the Albanese government expects will inject up to $600bn a year into the national economy by 2030.

But, Cisco ANZ’s director of cybersecurity Corien Vermaak said trust remains a big hurdle in fuelling AI’s widespread adoption.

Cisco ANZ head of cybersecurity Corien Vermaak.

Ms Vermaak said 27 per cent of Australian companies surveyed had banned the use of generative AI applications over privacy risks. Of that number, 77 per cent were worried corporate information entered into AI apps could be shared publicly or with competitors, while 69 per cent were concerned it could hurt their legal or intellectual property rights.

“As we do security awareness training, we now need to focus on that narrative on AI as well,” Ms Vermaak said.

“Where organisations are unsure, they’ve banned it. What that tells me is a lot of organisations are looking to put the controls in place, are looking to educate, message and see where they can improve their processes before allowing it. My sense would be over the next few years we will see that decreasing.”

While bigger companies have signed up to tailored enterprise AI products, like Microsoft’s Copilot — which has guardrails built in around protecting corporate data — others do not have a clear strategy, with employees at some using publicly available AI platforms, sometimes without their manager’s knowledge.

“One of the examples in our framework is that when an employee uses any generative AI tool, they have to actually disclose it. And, and that’s a really important thing, because it changes the perception of the receiver of that work,” Ms Vermaak said.

“There’s two things that organisations and companies need to start with. That is the education part, which is critically important. And then putting in policies and frameworks to put guardrails up for when is it okay to use it: ‘Can I use ChatGPT?’ ‘Yes, you can use it for private things on your work device, but you can’t use it for any work related matters’ if that is what the company’s risk acceptance is.”

The Productivity Commission also found low levels of public trust in AI technologies could pose a “significant barrier” to uptake. But, it said governments should act as “exemplars” to demonstrate the safe and effective uses of AI in service delivery.

Collete Stallbaumer, general manager of Microsoft 365, said it was still early days and companies were moving at different speeds in regards to AI adoption and their comfort with the technology.

“We take our responsibility very seriously and we are doing everything we can to ensure that AI benefits society broadly,” Ms Stallbaumer said.

“We’re just very early in this technology inflection point and it’s going to take time, it’s going to take education and it’s going to take people getting more and more comfortable with how to use to these tools. Microsoft’s approach is to be very engaged, whether that be with our customers or with governments around the world to ensure that AI is deployed safely and responsibly.”

Microsoft has found it takes about 14 months for companies to become proficient and reap benefits from AI. A study the tech titan commissioned from IDC found globally AI generated an average investment return of 3.5x every dollar spent. For Australian companies which had adopted AI systems, the return was higher, at 3.66x.