AI is here to stay, but where are its limits?

Though advice remains a human-centred profession, the role artificial intelligence (AI) is beginning to play within many firms is growing. Currently, AI is being used as a means of automating parts of the advice process, such as compliance checks and document generation, which in turn can be used to create efficiency and help when firms are looking to scale. However, some within the industry believe it can go further.

“I think one of the biggest areas of untapped potential for AI in financial advice is in delivering truly personalised, real-time planning for everyday clients – not just the ultra-wealthy,” Complete Wealth financial adviser Dr Ben Neilson told ifa.

“There’s a huge opportunity to use AI to provide tailored, goals-based guidance that adjusts dynamically as a client’s life evolves.”

According to the Tech Council of Australia’s (TCA) Future ready: Australians and AI workplace tech report, the general perception of AI in the broader Australian workforce is positive, with 93 per cent believing AI will augment, not replace their jobs.

“There is a willingness from the workforce to engage and augment their jobs with AI,” TCA chief executive Damian Kassabgi said.

“That’s not to say there is zero concern, the main concerns raised by Australians on adopting new technologies relate to privacy, training and being involved in future decision making. But it’s positive to see that overall, Australians are optimistic and open minded.”

However, as it stands, AI remains limited in its scope of what it can currently achieve.

“Right now, most tools are still fairly static – they’re good at creating a snapshot plan, but not necessarily at adapting in real time as spending patterns change, goals shift or unexpected events come up,” Neilson said.

Another major limitation on AI is its inability to fully grasp human emotion and the weight behind different decisions clients make when seeking advice. As many advisers know, though money is technical in nature, many reasons why people seek financial planning are emotional based.

Whether it be creating a legacy for one’s children or understanding if retirement in the country is reasonable, the emotional reasons for seeking advice can be hard for AI to quantify.

“It can’t fully grasp the emotional weight behind certain decisions – like when a client chooses to support a struggling family member over maximising retirement savings, or when legacy, guilt, or fear play a role in financial choices,” Neilson said.

“These aren’t spreadsheet decisions; they’re human ones. AI can only ever develop the process, it can't replace the human impact.”

Eric Blewitt, CEO of Investment Trends, echoed this sentiment, emphasising the importance clients place on human interaction.

“Trust remains human-led. Clients value empathy, judgement and accountability, areas where advisers continue to differentiate,” Blewitt said.

He added that “we anticipate that the winning model is one where advisers embrace AI to enhance their judgement, not replace it. The interpersonal skills of a trusted adviser remain critical”.

It is that client trust in the adviser and the need for an interpersonal relationship in the process that could also inhibit the integration of AI into the workplace. For Investment Trends, the key to this is centring clients into the integration process. This may look like:

- Providing transparency to clients about how AI is being used.

- Ensuring the “human touch” is evident throughout the process, reaffirming to clients that any advising decisions are being signed off by a human.

- Make controls and safeguards obvious to clients.

- Highlight the benefits of using AI, such as more face-to-face interactions.

Another key aspect in discussing AI’s limitations are the ones the industry should be placing on it. Like the sentiment shared across many industries grappling with AI, Neilson believes that the goal should be to enhance human advisers, not replace them.

“AI can help us scale, spot patterns in data we might miss and stay proactive instead of reactive,” He said.

For Kassabgi, government needs to step up and take a more proactive role in shaping policy and be seen doing so: “Our research found 49 per cent of Australian workers do not think government officials understand technology well enough to regulate it effectively. This perception needs to shift and that means government must lead by example.”

But within financial advice, Neilson argued that policy and understanding needs to be driven by professionals in the field, not lawmakers. However, he said as it stands, few advisers are having these discussions.

According to Neilson, the sector is “tiptoeing” around AI, often waiting on regulators and vendors to define boundaries for them.

“Practitioners should be the ones shaping how AI is trained, where it fits into the advice process and how it's audited for quality and bias If we want AI to augment – not dilute – professional judgement,” he said.

For Blewitt, human oversight and governance is key to ensuring AI is an effective tool.

“Just as organisations need to mitigate against ‘model risk’ where adverse outcomes can occur based upon incorrect or misunderstood data,” he said, those same processes need to be applied to AI.

“AI systems are prone to hallucination. It follows that human oversight and robust governance is required to mitigate against hallucination risk.”

With something as fast evolving as AI, it is unlikely that its integration will be standard across practices, with those willing to discover its limits and impose it themselves likely to be leaders in the future.

“The future of advice won’t wait for consensus. Those who lead with clarity and conviction around AI will shape the standards – and everyone else will just adapt to them,” Neilson added.