Artificial intelligence has been hailed as the most powerful technological leap since the creation of the internet. It can accelerate scientific discovery, transform medicine, improve government services, make businesses more productive and even help address global challenges such as climate modelling, disaster response and food security. Its promise extends to almost every corner of the planet.

Yet alongside the excitement sits a quieter, more urgent question: is anyone in danger from AI?

The answer, increasingly, is yes — but not because AI is inherently harmful. Rather, the danger emerges from how people choose to use, misuse or lose control of it. AI is a tool of immense capability, and like all powerful tools, it can uplift society or be turned against it.

Below, we examine who is most at risk and why.

The Everyday Consumer: Targeted by AI-Enhanced Scams

The average person is more exposed than they might think. AI has supercharged the sophistication of scams by allowing criminals to:

• Generate personalised phishing messages

• Clone voices for fraudulent phone calls

• Create realistic deepfake videos

• Mimic writing styles with startling accuracy

A scammer no longer needs to be tech-savvy. They simply need access to an AI model. That puts millions of everyday Australians — especially the elderly, the time-poor and the digitally inexperienced — at heightened risk.

Public Figures and Professionals: Vulnerable to Deepfakes

Politicians, business executives, celebrities, teachers, lawyers and even small-business owners are now more exposed to reputational harm than at any point in history.

Deepfakes allow bad actors to create fabricated audio or video that appears authentic. A false statement, a fake confession or a manufactured scandal could spread across social media before fact-checkers can react.

The danger here is broader than reputational damage — it affects journalism, trust in public institutions, market stability and social cohesion.

Workers in Automated Sectors: Disruption Without a Safety Net

Millions stand to benefit from AI tools that make work faster and more efficient. But for workers in industries heavily exposed to automation — transport, retail, customer service, logistics, administration and finance — AI poses real economic risk.

Jobs are unlikely to disappear overnight, but roles will shift, consolidate or require new skills. Those without access to retraining, or employers who fail to adapt, may find themselves sidelined by technology faster than expected.

The risk is not AI itself, but inequality — a widening gap between those who can adapt and those left behind.

Children and Teenagers: A Generation Growing Up With AI

Young people are in a uniquely vulnerable position. They live in a digital world shaped by:

• AI-driven content feeds

• Hyper-personalised advertising

• Online tutoring and chatbots

• Social media filters and deepfake tools

Children may struggle to distinguish between human-created and AI-created content, making them more susceptible to manipulation, harassment, identity theft and unrealistic social expectations.

AI can be transformative in education and learning, but without digital literacy programs, young people may face long-term psychological and social impacts.

Democracies: Threatened by Manipulation and Influence Campaigns

The integrity of elections — including in Australia — may face new threats from AI systems capable of generating:

• Large-scale political misinformation

• Fake social-media movements

• Hyper-targeted propaganda

• Synthetic news articles

• AI-driven bots that convincingly imitate real voters

Democracies are built on trust and informed decision-making. When AI tools are used to distort reality, the very foundation of democratic life is put at risk.

Businesses and National Infrastructure: Targets for AI-Assisted Cyberattacks

AI is making cybercrime more efficient and more dangerous. Companies, hospitals, utilities, and even government agencies are increasingly at risk from attacks using:

• AI-generated code to breach systems

• Automated vulnerability scanning

• Intelligent phishing schemes

• Ransomware that adapts in real time

Critical infrastructure — water, energy, transport, telecommunications — sits at the frontline. A successful attack could disrupt essential services and threaten public safety.

AI Developers Themselves: Facing Pressures They Can Barely Keep Up With

Ironically, those building AI are also at risk. Developers face pressure from:

• Commercial forces pushing rapid deployment

• Geopolitical competition

• Ethical scrutiny

• Unexpected system behaviours

• Potential misuse of open-source tools

Even well-intentioned creators cannot foresee every consequence of their models. As AI becomes more capable, the margin for error narrows.

So, Who Is Actually in Danger From AI? Everyone — But Not Equally

It is not accurate to say AI is “dangerous” in a vacuum. Most AI systems are safe, narrow and heavily regulated. The real risk lies in:

• Misuse by bad actors

• Vulnerabilities in human systems

• Lack of safeguards, oversight and digital literacy

• Rapid capability growth outpacing regulation

AI can help almost everyone on the planet. It can democratise knowledge, reduce inequality, improve healthcare, enhance safety and expand opportunity.

But without careful management, it can also amplify harm, accelerate crime, and destabilise systems that societies rely upon.

A Balanced Future Requires Smart Action Today

If AI is to fulfil its promise without unleashing unintended consequences, governments, developers, businesses and communities must work together to:

• Strengthen AI safety standards

• Improve public awareness and digital literacy

• Develop regulations built for rapid technological change

• Support workers through retraining and transition programs

• Invest in cybersecurity at a national level

• Promote ethical development and transparent oversight

AI’s potential is extraordinary. But potential alone is not destiny.

Human choices — not algorithms — will determine whether AI becomes humanity’s greatest tool or one of its greatest threats.

For now, the technology is poised to help billions. The responsibility lies with us to ensure that in doing so, it harms no one.