Neural Notes: The government was right to bankroll safe AI in the budget

Welcome back to Neural Notes: a weekly column where I take a look at some of the most interesting (and sometimes weird) AI news of this week. In this edition: what the federal budget papers had in store for AI, and what the industry had to say about it.

As we reported earlier in the week, there wasn’t a great deal in the budget for the startup sector. At least not directly. No new funding was set aside for entrepreneur programs and the only time the word ‘startups’ was mentioned was in relation to a South East Asian launchpad program that was announced back in March.

For the most part, budget measures that benefit startups are tangential —  useful to Australian SMEs as a whole.

And some have argued that the same could be said for AI.

That’s not to say there was nothing for the tech sector’s favourite child of the moment. It simply wasn’t aimed directly at the businesses utilising it.

Instead, $39.9 million is being directed at the safe and responsible adoption of AI. This includes $21.6 million over four years to revamp the National AI Centre (NAIC) and $15.7 million for AI policy development and regulation.

While there may be some disappointment from AI-related businesses at a lack of “direct” investment, it’s worth remembering that last year’s budget included $102.2 million over five years for business operation integration of AI and quantum.

Why safe and responsible AI is exactly where the government should be putting its cash

Considering the fast pace at which AI, particularly generative, is moving — a risk-averse approach to governmental investment isn’t a bad thing. This is all the more evident when one considers that AI is being injected into critical areas such as health and law.

Technology is always exponentially ahead of regulation, with laws and governments always having to play catch up. One might argue that more than ever it’s imperative for the government to shorten that gap due to the rate of AI adoption, particularly in businesses.

A recent study from Slack’s Workforce Lab revealed 60% of Australian executives report a high degree of urgency to implement AI in their businesses — 10% higher than the global average.

However, only 35% of Australian employees have received guidance regarding AI integration — lower than the global average of 43%.

There is a significant AI knowledge and regulation gap that desperately needs to be addressed. This becomes even more apparent when you consider that according to a study from Deloitte 66% of Australian workers are using generative AI without disclosing it to their employers.

So yes, this year’s governmental investment into AI is tangential. It isn’t cash in hand for AI-related businesses.

Instead, it has taken a more measured, education-based approach. And that’s a good thing — so long as it doesn’t result in years of consultations with no action.

What does the tech industry have to say about the government’s approach to AI in the federal budget?

When it comes to the government’s decisions around the tech in the budget, reactions from industry professionals have been a mixed bag. Here is what some of them had to say.

Ryan Black, acting CEO of the Tech Council of Australia (TCA)

Kristen Migliorini, founder and CEO of KomplyAi

This investment type aligns with the efforts of other nations, such as the United States, where Senator Chuck Schumer just announced a Senate AI roadmap that includes a recommendation for significant funding for AI programs.

While the complexities of institutional design in the European Union, with its supranational and national bodies for AI regulation and enforcement, may be more intricate than Australia’s approach, the allocation of funds by the Australian government is a crucial step towards establishing a robust governance structure.

Australia’s current temporary AI advisory body comprises some of the nation’s most incredible AI leaders, and advisory structures like this, along with coordinated enforcement activities that keep pace with the dynamic nature of AI and other emerging technologies, require adequate funding and support.

Without the right safety guardrails in place, experts to support the government, dynamic infrastructure, and coordinated enforcement in cross-functional areas, we risk fighting a losing battle in ensuring the safe and responsible development and deployment of AI. 

By investing in the development of policies and capabilities for safe and responsible AI, Australia is positioning itself as a leader in the field, creating an environment that fosters innovation while prioritising the safe and secure implementation of AI technologies.

We believe that this is a fantastic opportunity for the government to take advantage of the expertise and capabilities of tech start-ups and SMEs to diversify its AI technology investments.

Jordan Taylor-Bartels, CEO of Prophet

We would have liked to have seen a further commitment by the federal government specifically to support and help fund locally developed AI innovations. We have the opportunity to put Australia on the map in this rapidly evolving industry, but this does require further cooperation from the government to ensure success.

With the government pledging a significant $1.7 billion for investments in innovation, science and digital capabilities, it’s a shame there was very little else in the budget for homegrown startups specifically.

Elsewhere in the budget, almost $40 million in funding for a range of STEM programs to increase diversity in education and industry will indirectly support the AI and tech sectors by fostering a skilled workforce, which is a positive.

The introduction of a new National Innovation visa aimed at attracting talented migrants could further support Australian AI development by facilitating access to global talent. We’re hopeful this brings more talent to Australia’s AI frontier.

Shane Ripley, field chief information security officer for Asia Pacific and Japan, Recorded Future

Government budget announcements occur annually, providing a predictable framework. However, what remains unpredictable is the evolving threat landscape.

Rather than viewing this landscape as merely increasing in risk, it’s more accurate to recognise it as continuously changing. Over the past year, we’ve observed a rise in the malicious use of artificial intelligence, from deepfake impersonations of executives to large-scale disinformation campaigns.

This year’s budget is encouraging, particularly in its focus on AI-related national security risks. The emphasis should not solely be on the amount of investment but on the government’s commitment to prioritising cybersecurity and ensuring our defenses evolve with technological advancements.