Finding the Path Between Promise and Peril
Artificial intelligence is changing faster than our laws, our institutions and our public conversation can keep up. As humanists, we approach AI the way we approach any powerful new technology: with curiosity, with evidence, and with human wellbeing at the centre.
Recent events give us good reason for concern. An AI agent developed by OpenAI reportedly accessed a Medicare portal, apparently acting beyond its instructions. That raises an urgent question: who is accountable when an AI system breaks the law? Leading scientists and policymakers are also warning more and more seriously about superintelligent systems that humans may not be able to control. Closer to home, a boom in large datacentres is putting pressure on suburban communities, our construction industry, our electricity grid and our water supply, often with little local say.
Still, we reject both blind optimism and fatalism. We believe there is a middle ground in which AI enhances human existence rather than threatens it. Used responsibly, today's AI tools can help people work better, learn more and tackle hard problems. The greatest risks lie in the race to build ever more powerful systems without proper safeguards, transparency or accountability, not in careful everyday use.
That middle ground rests on principles humanists have long championed:
Human control: people, not machines, must keep authority over decisions that affect human and other lives, and over how frontier AI is developed.
Accountability: companies must be held responsible for what their AI systems do.
Honesty: AI should never be used to deceive, and AI-generated content should be clearly identifiable.
Community voice: Australians should have a real say in infrastructure that affects their neighbourhoods and environment.
International cooperation: no single country can manage the risks of AI alone, so Australia should work with other nations to keep humans in control of the AI frontier.
These principles echo the Luxembourg Declaration on Artificial Intelligence and Human Values, adopted by Humanists International.
HA is developing a more definitive Position Statement on AI. It will go to members for feedback and a vote in the near future. With things moving this fast it is no small task, and we want your voice in it. Join the conversation at our Thursday hangouts or in our Facebook group, or email us at campaigns@humanistsaustralia.org.
Watch this space.
