Despite heavy investment in digital safety systems, artificial intelligence, and integrated compliance platforms, Australian workplaces are struggling to translate spending into safer outcomes on the ground, according to new national research.
The report commissioned by Rapid Global and conducted by Research Without Barriers, surveyed more than 1,000 safety managers, workers, and contractors across high-risk industries. Its findings reveal a widening disconnect between leadership confidence and frontline reality, raising concerns that safety is becoming well documented but inconsistently experienced.
While nearly two-thirds of workers say safety processes are clear and practical, only 41 per cent believe safety is taken seriously by everyone, all the time. The gap, experts warn, reflects a deeper issue of trust between policy and practice.
“Safety is often well documented, yet not consistently felt by people on the ground,” says Professor Dr Andrew Sharman, global authority on safety culture and CEO of the International Institute of Leadership & Safety Culture. “Bridging that gap is less about systems alone and much more about leadership. Trust is the critical differentiator.”
The research highlights usability as a major fault line. Two-thirds of managers say unifying safety systems would simplify compliance, yet fewer than half of executives believe their current tools use modern, easy-to-use technology. Among frontline workers, just 41 per cent find safety software easy to understand, a figure that drops sharply to 30 per cent in larger organisations.
When systems are difficult to use, workers often bypass reporting or rely on manual workarounds, creating blind spots that undermine risk management. Nearly one in four workers say they have personally seen incidents go unreported, while half of managers admit incident reporting could be simpler.
Despite years of digitisation, paper-based processes persist. Almost a quarter of Australian businesses still rely on paper for critical safety tasks, particularly in industries with remote sites and large contractor workforces. Even where digital tools exist, friction remains a barrier to consistent adoption.
Artificial intelligence is emerging as another point of tension. While 64 per cent of managers believe AI and robotics will fundamentally transform workplace safety within five years, only 25 per cent of workers share that optimism. At the same time, 41 per cent of managers admit they are already using unofficial AI tools to assist with safety tasks, suggesting demand is moving faster than governance.
“Technology should not replace human judgement but make it sharper,” says Ezequiel Gonzalez, Head of Revenue at Rapid Global. “Complex, high-risk environments require more than ticking boxes. When systems are easier to use and data is easier to act on, safer outcomes follow.”
Even among managers, enthusiasm for AI is tempered by caution. Six in ten say AI should support analysis rather than make safety decisions, reinforcing the central role of human oversight in high-risk settings. Adoption is also uneven across the country, with organisations in New South Wales, Queensland and Western Australia more than twice as likely to use AI-enabled safety systems than those in Victoria and South Australia.
The report also flags weaknesses at the point of entry. More than one-third of workers and managers say it is still possible to access worksites with incomplete or expired training, despite broad agreement that linking induction data directly to site access systems would significantly improve safety.
According to the research, organisations most likely to see real improvements are not those adding more tools, but those reducing friction, automating enforcement, and making safe behaviour the easiest option on site.
For Australian leaders, closing the gap between strategic intent and frontline experience may be the single most important step in preventing small, systemic cracks from turning into serious workplace risk.