After a recent MIT report found that 95% of AI projects fail to deliver a return on investment, new research found that a growing number of Australians are negative about AI.
The latest Roy Morgan poll found that 65% of Australians believe AI ‘creates more problems than it solves and 25% believes that AI presents a risk of human extinction within the next two decades.
This is a really significant finding for institutions which have been burnishing their AI credentials over the past 12 months.
The data shows that while tertiary institutions are rushing to demonstrate their AI credentials, there is a fair way to go to win over a sceptical public.
The survey of 3543 Australians conducted at the start of this month found that women, older people and regional residents tended to be more negative about AI, while South Australia is the State with the most non-believers in the technology.
Amongst the 35% of Australians who believed AI presented a net positive, reasons for embracing the technology included improved productivity, access to knowledge and a belief, borrowing from the NRA’s pro-gun mantra – “AI doesn’t create problems, people do”.
The fact that one in four respondents believed that humans may be extinct by 2045 as a result of AI reflects a surprisingly widespread deep-rooted suspicion of the technology.
The concerns may be fed by research such as the recent MIT study, which found that many businesses hoped for sales and marketing breakthroughs with AI bots, discounting the value of humans in the acquisition process, and back-office automations delivered the highest returns – streamlining processes and cutting costs.
As HE institutions move beyond the gee whiz towards the ‘What’s Best’ strategy in 2026, it is clear that a lot of work is required in shaping internal frameworks to extract value and external positioning to carefully calibrate exposure to negative sentiment.