Australia risks being in a position where it has little say on the AI systems that will increasingly affect its future. An Australian AI Safety Institute would allow Australia to participate on the world stage in guiding this critical technology that affects us all.

Declining trust across society is a barrier to widespread adoption of AI in the community, and distrust in AI due to misuse will further weaken Australia's democratic institutions.

Decisions about revolutionary technologies, like AI, should be made democratically.

I support the establishment of an Australian AI Safety Institute and the introduction of an AI Act. We don’t want advanced AI that is unsafe, untrustworthy, or unreliable—no one is better off in that scenario. Unfortunately, that may be what we are racing toward.

As someone who has worked across industry, research, and national infrastructure, with deep experience in AI systems, socio-technical modelling, and AI applications and risk, I see the establishment of an AI Safety Institute as a critical step for Australia. AI systems are increasingly being integrated into society. AI is not just technical but are socio-technical, shaped by human values, institutional processes, and data infrastructures. Australia needs an AI Safety Institute to guide, assess, and coordinate safe, ethical, and high-impact AI development. The institute should be framed around socio-technical considerations, and should bridge policy and practice in areas like research data infrastructure, model validation, and governance, provide guidance on responsible AI design and deployment, align with democratic values, Indigenous data sovereignty, and public trust, and offer a platform for multi-stakeholder dialogue. With Australia investing in national research infrastructure and expanding its AI capabilities, it is crucial to lead in ensuring safe, inclusive, and transparent systems for both current and future applications. The institute should connect policy formulation with implementation, provide guidance to public institutions, and create opportunities for diverse stakeholder engagement.

We need to ensure AI systems align with our society’s needs and values, while ensuring a populace with a healthy educated skepticism of these systems. This is best achieved through an Australian AI Safety Institute and AI Act.

I support the creation of an Australian AI safety institute and the implementation of an AI act as both are urgently required to ensure that the risks associated with AI are effectively managed. We are fast losing the opportunity to ensure that all AI technologies are safe, ethical, and beneficial to humanity.

Intelligence shapes the world. We have the responsibility to design systems that extend not only our intelligence and creativity, but also our wisdom and conscience. AI should be deeply ethical by design.

I have done research with AI tools and technologies in terms of understanding their bias, training data, and potential. I believe AI safety is important given the rapid proliferation of these tools and technologies in our everyday lives.

An AI Safety Center will allow Australia to coordinate and accelerate our disparate efforts in AI Safety and Ethics, paving the way for regional leadership in this strategic and important area.

AI is as transformative as electricity and as powerful as nuclear technology. We wouldn’t handle those without clear mandatory safeguards, and AI should be no different. To support good policy, Australia's government also needs a dedicated AI Safety Institute to bring deep technical AI expertise into government.

AI is the test case for how we deal with every one of the seismic changes of the 21st century.

It sets a dangerous precedent for Australia to formally commit to specific actions but fail to follow through. Australia is the only signatory that is yet to meet its obligations by creating an AISI.

AI safety is crucial for Australia’s threat landscape. As I have been doing for the last few years, I strongly advocate for social scientists to be included at the forefront of this movement.

Working in London's tech cluster, near DeepMind and Google, gives a front-row seat to AI's rapid evolution. In 2019, my DPhil research on AI for detecting mental health issues on social media met blank stares or dismissal. Two years later, ChatGPT arrived, and those skeptics stopped laughing. AI's pace of change makes prediction nearly impossible. For these trailblazing companies, 18-24 months is the far future. While committees deliberate and leaders pontificate, developers in their early twenties create autonomous agents in hours and write code in seconds. This happens in countless apartments and incubators right now. Anthropic's CEO Dario Amodei recently spoke at global forums about AI scenarios that should give us pause. Not science fiction, but plausible futures where systems might operate beyond our understanding. Nuclear weapons have predictable effects. AI systems may not offer such clarity—operating in various planes, conducting operations against opposing AIs. Gradual escalation would be almost inevitable as computer logic leaves no room for empathy. What concerned me most wasn't worst-case scenarios, but Amodei's warnings about jobs and economic security. The coming wave won't just hit specialized roles. Potentially vast segments of our economy will cease to require human assistance. We see early signs. Freelancers in creative and technical fields are our canaries, watching tools emerge that do in seconds what experts took days. AI agents don't sleep, don't need benefits, and improve rapidly. No sector will remain untouched—whether you work with hands or mind. Most countries, including Australia, are unprepared. Policy discussions crawl while governments hide behind inquiries that produce reports and little else. Meanwhile, the gap widens daily. This change won't arrive as a dramatic collapse, but as a quiet tide reshaping our economic shoreline. What happens when your expertise becomes obsolete? What about your family? Your community? Imagine: what would you do if you couldn't rely on your job, with no clear path to employment? When tech companies race to launch advanced projects quarterly, when experts speak with increasing concern—it's worth paying attention. This isn't about preventing progress or succumbing to fear. It's about recognizing that significant change is coming, likely sooner than we think. The window for thoughtful action remains ajar, but won't stay open indefinitely.

One of the biggest mistakes in rapidly evolving situations is to assume the present defines the future, and our understanding of today's situation is sufficient to navigate what is unfolding. We must recognise that failing to research and understand advancing AI developments means we, as a nation, will be leaving our wellbeing in the hands of others who may care little. An Australian AI Safety Institute would help safeguard the nation by providing the understanding needed to make informed decisions about AI adoption in our unique nation. We are unable to express a preference if we do not understand the subject. The best time to plant this tree was 10 years ago. The second best time is now.

As the CEO of the Humanitarian Assistance and Disaster Relief Institute, and as a researcher exploring the application of international law to AI Decision Support Systems in armed conflict, it's clear to me that these technologies offer immense opportunities alongside significant risks of harm. Navigating this delicate balance demands active, innovative governance and responsible leadership.

Robust assurance justifies trust. We're all excited about the potential opportunities of AI, but not enough work is currently happening to address genuine safety concerns. It's easy to understand why Australians are hesitant to adopt AI while these big issues are outstanding.

I believe -- along with many of my colleagues researching artificial intelligence -- that the 48th parliament of Australia may preside over an intelligence explosion in which human beings will no longer be the most intelligent beings on the planet. I commit to doing whatever I can to assist my local member and any other member of parliament in understanding the issues at hand and to offer whatever guidance I can.

I support the calls for action of Australians for AI Safety to assure citizens are protected from the downside risks of automated, algorithmic and other technologies and 'AI'.

Once something is smarter than you, it's too late. Before then, we need to make sure ASI robustly shares our values and won't enable totalitarian control.

AI safety and security is a global challenge. But middle powers like Australia have an important role in shaping the global discourse and strengthening safety measures.

AI safety matters to me because it sits at the intersection of my professional mission and personal values. As someone who designs and deploys AI solutions for public services, I’ve seen firsthand how the benefits of AI can be undermined by unintended consequences. Ensuring that the systems I build are safe, transparent, and aligned with human values is not just best practice, it’s a responsibility I owe to the communities I serve. Impact on society: AI-driven decisions now influence healthcare outcomes, justice processes, infrastructure planning and more. Without rigorous safety guardrails, biases and errors can amplify harm at scale, eroding trust in institutions and technology alike. Professional integrity: Having delivered over 500 hours of Data & AI workshops for government teams, I know that trust is earned through transparency and accountability. Embedding safety frameworks into every AI project reinforces that trust, ensuring stakeholders feel confident in adopting new solutions. Ethical stewardship: My passion for AI governance drives me to balance innovation with oversight. By championing policies and practices that anticipate risks, rather than simply reacting to them, I help steer AI toward outcomes that serve the common good. Community engagement: Hosting the Perth Machine Learning Group reminds me daily of the power of diverse perspectives. Inclusive safety practices not only catch blind spots but also democratise AI benefits across all segments of society. Long‑term viability: True innovation flourishes only when people feel safe using it. Prioritising AI safety today lays the groundwork for sustainable adoption tomorrow, unlocking human potential without compromising our ethical choices. By making AI safety central to my work, I strive to build resilient systems that empower users, protect vulnerable populations, and uphold the values that guide me - because a future worth building demands nothing less.

As artificial intelligence continues to evolve, the importance of AI safety grows exponentially. AI has the potential to enhance society in profound ways, but without careful oversight, its risks could outweigh its benefits. I believe that responsible AI development must be grounded in rigorous research, ethical considerations, and proactive safety measures. By signing this open letter, I am advocating for increased attention to AI safety within the research community. It is imperative that we prioritize transparency, fairness, and accountability in AI systems to safeguard against unintended consequences. Future advancements should not come at the expense of human values, privacy, or security. I urge researchers, policymakers, and industry leaders to come together in ensuring AI is developed with safety at its core. Only through collective responsibility can we build AI that is both innovative and safe for humanity.

I support government and industry taking strong action on AI safety, but I'm neutral on the exact form that will take. Any equivalent of an AI Safety institute is fine with me!

AI is presently impacting our lives and it will become a real concern for the future of the coming generations. We need to legislate it before it is too late!

AI is the issue of our time. Technology is transforming the world faster and faster - we must keep up to prevent great harms.

AI Safety is the key to keeping Australian medical data safe from unauthorized abuse in the medical insurance industry.

AI is getting increasingly capable, approaching or exceeding human performance on many real-world tasks, but our scientific understanding of how these systems work is remarkably nonexistent. Powerful technology that we understand poorly is a recipe for disaster. We need to act to fix this before it's too late.

Generative AI poses the greatest threat the creative sector has ever seen. In addition to this, the risk to democracy and truth is a price every Australian will pay if AI-generated media is not regulated.

With the advancement of AI, there are psychological consequences for the well-being of individuals as well as a significant impact on societies. Human work will continue to be transformed and will possibly be eliminated in the not so distant future. Interfaces that directly connect the brain with the internet will have an impact on how we think and communicate. The decisions and actions of an advanced form of artificial intelligence will be more and more difficult to understand, and hence, better forms of explanation for artificial intelligence are required. The technology is increasingly being used to manage significant parts of society, e.g. by use of social credit systems, with consequences for the entire population. Finally, advancements in military AI may include autonomous killing machines that can spread fear and terror. All these developments are happening as we speak and represent significant challenges to human psychological well-being. Advanced forms of artificial intelligence will have an impact on everybody: The developers and users of AI systems as well as individuals who have no direct contact with this form of technology. This is due to the soliciting nature of artificial intelligence: AI wants to be used and the "universal solicitation" of the technology is a challenge. We need an AI that is not just “human controlled” but beneficial in the sense that it explains itself and its operation to everybody. This includes the most vulnerable in a society, including children, the elderly and persons with an intellectual disability. Living with an artificial superintelligence is a critical area of research and resources should be allocated to it to help safeguard human existence as we know it.

AI safety is essential to both our national security and sovereign economic interests. I support a strong, coordinated approach - anchored by an AI Safety Institute and mandatory guardrails - to ensure safe, responsible AI development to support safe, responsible AI deployment. As a late adopter, Australia can now draw on global lessons - enabling the government to lead with clarity, embed sector-specific best practices and unlock AI’s full potential to strengthen our economy and way of life.

AI systems are rapidly gaining capability and are very likely to transform the world in the next few years. We must do everything we can to ensure the changes are positive.

While promoting innovations, risks should also be managed effectively.

We are building increasingly intelligent systems, but our methods of aligning their goals with ours are rudimentary and may not scale. Public funding to research this technical problem is essential if we wish future AI systems to remain safe.

AI is a fundamental net-negative to society on various fronts, whether it be its upward pressures on scams, downwards pressure on critical thinking, or the outright replacement of various skills. Effective policing at the policy level will enable the public and private sectors to create solutions that mitigate the potential damages autonomous and AI-assisted systems will cause.

Almost overnight we've arrived into a world where AI capabilities present serious risks to the safety, health and fundamental rights of Australian citizens, with little to no rules steering this technology in the right direction. The Australian government must act now.

AI is wonderful technology but absolutely needs regulation, and if we don't act now, we will be at risk of inundation of misinformation, false information, deepfakes and societal manipulation on scales that are unimaginable. The security implications alone are terrifying, let alone the risks of societal division and societal takeover by malign forces. We need to act constructively, together.

People need to be at the centre of any AI initiative

Even if you don't think AIs are going to enslave humanity, there are plenty of potential dangers, including the loss of jobs, the spread of misinformation, and the automation of warfare, to name just a few. We need to go into the AI era with our eyes open.

AI has arrived in waves, and no wave has delivered what it promised. This current wave, however, goes beyond those in the past. It offers management and decision makers the chance to bypass experts. There is a massive risk to sovereignty since the tools and their training sets are being supplied by sources in other countries who may seek to influence our domestic politics. Policy decisions made with the use of these tools can easily be tainted to achieve outcomes sought by foreign countries. An obvious example is Australia's various plans to control social media. These tools could permit covert influence being brought to bear via false data and conclusions.

Society is rapidly becoming reliant on AI without due consideration of the consequences.

It seems absolutely possible that smarter than human AI will be developed in the next 5 to 10 years. This will not be a new technology; it will be a new form of life. We will wish we had to put regulations in place when we still had the chance. Politicians will regret not having done more.

We are all beneficiaries of science and technology and the high standard of living they have created. Technological advancement is a wonderful thing, but when working with new technologies that are changing rapidly and have emergent properties that are difficult to interpret, caution and public consultation are essential. Australia has a moral obligation to meet its commitments regarding AI safety, and an opportunity to be a global leader in the responsible development of AI rather than merely reacting to developments as they happen. We should take this opportunity.

Without government intervention, we're entrusting humanity's future to a few profit-driven companies developing artificial intelligences they barely understand, effectively giving these corporations and a small group of foreign officials permanent control over humanity's future.

I am hopeful that as AI technology matures it will contribute significantly towards improving people’s quality of life while being implemented in an ethical and responsible manner; however, the ethical use of AI cannot rely on the altruism and goodwill of individuals. Currently, day-to-day experiences of the limitations of AI, and observations of its misuse, have raised concerns around data integrity, privacy and its potential to cause harm. AI by itself is neither good nor bad; the purpose for which it is used and the way it is implemented determines that characteristic. The role of well-developed policies, standards, regulations, and governance framework in the risk management and use of AI will determine its overall benefit or harm to society.

AI presents a very real threat of "identity theft" of the highest order. Not simply usernames or password, but the very essence of what make "me" me. My appearance, my mannerisms, my voice and... to the observer, my message. A lack of checks and balances in this sphere threatens the potential for a lawless, parallel universe, where reality and "make believe challenge" each other for an audience attention and where influence can be brought to bear by those whose motives are suspect.

The future of AI is critical to the development of the human race. It is essential that we take the time to get this right and to ensure appropriate safeguards are in place.

It is our largest opportunity to humanity but also the most significant risk to humanity. Understanding the risks, being transparent, active management and mitigation is the only way that AI can be sustainable in the future. AI risk is not a barrier but an enabler. This is what the world needs to understand and back.

I am a Master’s student at UNSW, studying AI. I am doing this specifically with the intent to work on research relating to technical AI safety. I’m also completing the Technical Alignment Research Accelerator in Sydney. My background is in mathematics and physics and I have chosen to alter my career path because I believe, based on the evidence available and the arguments made by leading experts, that AI advancements will be the most consequential events of our time, and that without policy driven safety measures we will be choosing to use the most powerful technology ever invented without any consideration for the most dire consequences. We must be prepared for this technology.

As someone who has closely followed AI development for years, I'm signing this open letter because I believe we stand at a critical juncture. The rapid advancement of AI technologies presents both extraordinary opportunities and unprecedented risks that demand thoughtful governance. My professional experience has given me insight into how transformative technologies can outpace regulatory frameworks. With AI, this gap is particularly concerning given the technology's potential to fundamentally reshape our economy, society, and democratic institutions.

I 100% support AI Safety. AI is evolving faster than anything we've seen—it's powerful, but dangerous if left unchecked. Entire industries will be disrupted. Jobs lost. Lives changed. We must plan for this, not react to it. Artificial Superintelligence is no longer science fiction—it’s on the horizon. We need strong, transparent, global safeguards now. AI must serve people—not control them.

AI must be used for good only. We must have guardrails so as to protect ourselves against rogue actors who would chose to misuse it for power.

AI presents unimaginable emergent risks and will utterly transform society and cause the worst harm ever to the natural world, without careful regulation and oversight.

The development of artificial intelligence is momentous, with implications on a civilizational scale. The contents of this letter are not radical; rather, they represent a baseline that any responsible governance structure should adopt as the bare minimum standard.

The right regulations will enable Australian businesses to innovate. The safety institute is needed to provide the much needed interpretative guidance to businesses.

As an actuary I recognise that businesses and consumers both greatly benefit from clarity around responsibility for risk, and the value of supporting businesses in managing emerging risks. This is why I believe the Australian government must take action on AI Risk now.

I'm concerned that there aren't enough safeguards. I'm also concerned about students and professionals relying on AI to do their research for them and them not doing enough themselves. Most particularly in the fields of science , medicine and law. ALL fields that require absolute correctness so that there is no risk of e.g. incorrect medication, insufficient or erronerous data ( in scientific areas) or incorrect judgments or findings in the law from ill researched information and case law. I realise that AI is SUPPOSED to work on information it receives, but that presupposes that information is pure and unbiased

- I am concerned AI technology will develop without governance in a similar way to the advent of social media - I would like technology companies to be bound by a duty of care - I am appalled by the theft of original work by AI models - Whilst I acknowledge there are many benefits that may come with AI, I am concerned that the energy used by AI applications will make emissions targets unattainable

It's impact on our lives is going to only grow so we need to make sure that it is aligned with humanity's best interests.

Australians deserve the safe application of technologies that may radically change our world or pose unforeseen risks.

We should safeguard a powerful dual-use technology like AI, both nationally and internationally.

Over the last 30 years in particular, global inter connectivity has empowered state and non-state actors to deliver dangerous and deadly effects of all types across the world. AI will make this better or worse - we need to act now to make sure it is the former.

To make use of AI and really progress society for the better, we need to be able to trust it and know it won't compromise our safety. There is not enough transparency for us to know how AI tools will be used by each government and how we'll be protected by criminals.

I have long been optimistic about the future of humanity, and have watched in awe as our technology has progressed over my lifetime. However, the rapid increase in the capabilities of artificial intelligence has me afraid. With no means to control an entity significantly more intelligent than us, the chances of a positive future for humanity reduce day by day. All the many young people I have taught over the last decade are at risk, and I am terrified that what I promised for them - the potential for a wonderful future - is going to turn into a nightmare and then into nothing. For the sake of them and for all the young people like them across the globe, we must act. And act immediately.

AI safety matters for the future of all of our society and the generations to follow. To have a flourishing future we need to roll out AI in a way where safety is at the forefront, not an afterthought.

I regularly use AI, and LLM's for my work in the renewable energy field and see first hand tremendous opportunities and use cases. However, the more powerful it becomes, the more it is obvious to me that there must be strong standards to ensure public safety.

I believe AI can be a force for great good in the world but current economic and sociopolitical pressures are creating an environment wherein AI will be utilised not for the good of all but for the good of very, very few-- and the decisions of those few will decide the priorities of AI development, priorities unlikely to include any real commitment to public safety over 'progress'.

Imagine you are speeding down a winding, perilous road and a passenger says "Look, let's just slow down a bit. It's important we get there safely." That's AI Safety. They continue, "Wait, you guys aren't wearing seatbelts? Do you even know where we are going?"

AI has the potential to utterly change the world, and it is essential that we as a society direct that change to be beneficial rather than just hoping things turn out well by accident. Australia has the potential to be a thought leader in this field - if we come up with good policy here, we can be an example to all the other nations yet to legislate on the issue at all, which notably includes the US federal government.

AI safety is the most important issue to me this election. AI presents real risks to Australia's security, economy and the jobs, health and wellbeing of all Australians. Australia's leaders need to take a proactive stance to ensure that AI is safe and beneficial to humans, both here and on the world stage.

AI is going to have a future altering impact on society, whether that is positive or negative is finely in the balance. Progress is rapid, now is the time to think about safety before it becomes too late.

For the future of humanity