Category: MIL-OSI

Source: The Conversation (Au and NZ) – By Benjamin Liu, Senior Lecturer in Commercial Law, University of Auckland, Waipapa Taumata Rau

Most of us would agree fairness is a good guiding principle in life. Actually defining and applying it in the law, however, isn’t quite so simple.

Since March last year, New Zealand’s financial sector – including banks, insurers and credit unions – has been governed by the Conduct of Financial Institutions regime.

At its centre sits a principle that “financial institutions must treat consumers fairly”. Under the Financial Markets Conduct Act 2013 (and amendments made in 2022), the regime is administered and enforced by the Financial Markets Authority.

Each financial institution must establish, maintain and publish a fair-conduct program that satisfies a set of statutory minimum requirements.

These prescribe internal systems, controls, monitoring and governance processes intended to demonstrate the institution treats consumers fairly in practice. Breaches can incur a “pecuniary penalty order”.

On its face, this is uncontroversial. Fairness offers moral comfort and signals decency and responsibility. But translating fairness into a legal obligation is not without cost.

It also risks compromising consumer autonomy and informed choice by forcing financial institutions to limit the shape or scope of products and services that might otherwise be attractive.

Subjective regulation

While section 446C of the act provides broad definitions of fair treatment, it leaves significant scope for interpretation by regulators and institutions.

The result is a regulatory model that is essentially subjective and which shapes the design and distribution of financial products before they go to market.

This presents practical challenges for intuitions adapting to a fairness standard that is inherently vague. But it also raises questions about the balance between consumer protection and potential regulatory overreach.

In 2024, the government consulted on whether the statutory minimum requirements for fair conduct programs should be repealed or amended.

This was in response to industry concerns that some fairness requirements were either unnecessary or duplicated other regulations, or they were unduly prescriptive given the actual risks of harm to consumers.

Industry submissions generally acknowledged the high compliance costs associated with the current framework while supporting the broader objective of fair consumer treatment.

In response, the government chose to amend rather than repeal those minimum fairness requirements. In 2025, it introduced a draft amendment bill proposing changes to the statutory requirements for fair conduct programs.

If enacted, this may make the regime less strict. But it would also force institutions that have already invested heavily in compliance under the existing law to review and modify their programs once again.

Unintended consequences

This revisiting of the law reflects the the difficulty of defining fairness as a legally enforceable standard. Fairness is not an objective concept. It’s subjective and evaluative. What’s fair to one person may not be fair to another.

Yet the law now requires that financial institutions effectively prove they are designing and offering products and services in ways that align with the Financial Markets Authority’s evolving understanding of fair treatment.

As a result, even where consumers understand a product’s features and willingly accept its risks, the fairness obligation may still require institutions to reconsider whether the product should be offered at all.

On the surface, prioritising consumer interests over consumer choice might seem reasonable. But it can have unintended consequences.

In 2021, for example, the government amended the Credit Contracts and
Consumer Finance Act to impose highly prescriptive affordability checks on all consumer lending.

A 2022 investigation by the Ministry of Business, Innovation and Employment found the reforms had caused borrowers who should have passed the affordability test were being declined or offered reduced credit.

Fairness and risk

Because the fairness principle is broad and subjective, even if the Financial Markets Authority’s current interpretation is reasonable there is no guarantee future enforcement will be.

Once parliament embeds an open-ended moral concept in law, it hands significant discretion to whoever interprets it next.

Of course fairness matters. But it should be a moral compass for financial institutions and a cultural expectation for financial markets rather than an opaque licence for regulatory paternalism.

It risks turning financial institutions into overseers of consumer behaviour rather than providers of products and services.

It would be more straightforward to enforce existing laws such as the Credit Contracts and Consumer Finance Act and the fair-dealing provisions in the Financial Markets Conduct Act.

The aim should be to target specific misconduct, strengthen consumers’ financial literacy through education, and intervene where there is genuine, demonstrated harm.

The law should preserve the space for consumers to make their own decisions, even when those decisions involve risk. Fairness is a virtue, autonomy is a right. We should be careful not to sacrifice the second in the name of the first.

Benjamin Liu does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.

– ref. NZ’s finance industry is required by law to treat customers fairly – but how do we define ‘fair’? – https://theconversation.com/nzs-finance-industry-is-required-by-law-to-treat-customers-fairly-but-how-do-we-define-fair-272413

Evening Report: https://eveningreport.nz/2026/01/30/nzs-finance-industry-is-required-by-law-to-treat-customers-fairly-but-how-do-we-define-fair-272413/

Source: The Conversation (Au and NZ) – By Allen Cheng, Professor of Infectious Diseases, Monash University

An outbreak of the deadly Nipah virus in India has put many countries in Asia on high alert, given the fatality rate in humans can be between 40% and 75%. Several countries, including Thailand, Malaysia and Singapore, have introduced new screening and testing measures, after at least two people died of Nipah virus in the Indian state of West Bengal this month.

But what is Nipah virus, and how concerned should we be?

Here’s what you need to know.

What is Nipah virus?

Like Hendra virus, Nipah is in a category of viruses called henipaviruses. It is zoonotic, meaning it can spread from animals to humans.

As I explained in a previous Conversation article, outbreaks happen in Asia from time to time. The first outbreak was reported in 1998 in Malaysia.

There are three major ways it’s transmitted.

The first is via exposure to bats, and in particular via contact with the saliva, urine or faeces of an infected bat. Infections can also occur from contact with other infected animals, such as pigs in the original outbreak in Malaysia.

The second way it can be transferred is by contaminated foods, particularly date palm products. This means consuming date palm juice or sap that is contaminated with the bodily fluids of infected bats.

The third is human-to-human transmission. Nipah transmission between humans has been reported via close contact such caring for a sick person. This can mean, for instance, being infected with bodily secretions contaminated with the virus in households or hospitals. This is thought to be less common than the other transmission pathways.

What are the symptoms?

Nipah virus infections happen quickly. The time from infection to symptoms appearing is generally from four days to three weeks.

It’s a terrible disease. Around half the people who get severe Nipah virus infection die of it.

The symptoms can vary in severity. It can cause pneumonia, just as COVID could.

But the illness we worry most about is neurological symptoms; Nipah can cause encephalitis, which is inflammation of the brain.

These effects on the brain are why the fatality rate is so high.

Symptoms might include:

• fever
• seizures
• difficulty breathing
• falling unconscious
• severe headaches
• being unable to move a limb
• jerky movements
• personality changes, such as suddenly behaving oddly or psychosis.

Unusually, some patients who survive the acute phase of a Nipah infection can get relapsed encephalitis many years later, even more than a decade later.

Is there any treatment or vaccine?

Not yet, but in Australia development of a treatment called m102.4 is underway.

There was a phase 1 trial of this treatment published in 2020, which is where researchers give it to healthy people to see how it goes and if there are any side effects.

The trial found that a single dose of the treatment was well tolerated by patients.

So it is still quite a way off being actually available to help people infected with Nipah virus, but there’s hope.

There is currently no vaccine for Nipah virus. In theory, m102.4 it could be a preventative but it’s too early to say; at this point it is being trialled as a treatment.

How worried should I be?

This Nipah outbreak in India is worrying because there’s currently no prevention and no treatment available, and it’s a severe disease. While it is an important disease, it isn’t likely to be a public health issue on the same scale as COVID.

This is because it doesn’t transmit efficiently from person to person, and the main way it is transmitted is from food and infected animals.

For people living outside of areas where cases are currently being reported, the risk is low. Even in the affected areas, the number of cases is small at this stage, but public health authorities are taking appropriate control measures.

If you become unwell after travelling to areas where cases have been reported, you should let your doctor know where and when you travelled.

If someone gets a fever after travelling to affected areas, we would probably be much more worried it was caused by other infections such as malaria or typhoid than Nipah, at this stage.

Overall, though, everything needs to be put in context. We hear about new viruses and incidents all the time. Nipah is important for affected countries, but outside of those countries, it is just something we closely monitor and be alert for.

Allen Cheng receives funding from the National Health and Medical Research Council and the Australian Department of Health and Aged Care, including for public health surveillance systems. He has been a member of the Australian Technical Advisory Group on Immunisation and the Advisory Committee for Vaccines.

– ref. What is Nipah virus? And what makes it so deadly? – https://theconversation.com/what-is-nipah-virus-and-what-makes-it-so-deadly-274725

Evening Report: https://eveningreport.nz/2026/01/30/what-is-nipah-virus-and-what-makes-it-so-deadly-274725/

Source: The Conversation (Au and NZ) – By Ian Kemish, Adjunct Professor, School of Historical and Philosophical Inquiry, The University of Queensland

Canadian Prime Minister Mark Carney’s Davos speech should unsettle Australian strategic thinkers, who have been raised in the belief the US alliance is the unshakeable foundation of Australia’s regional security.

Carney’s point – that American leadership is no longer a reliable anchor for the international system – had strong appeal in Europe and Canada. But it also highlights what is now clearly the weakest link in the US-Australia alliance – not American capability, but American reliability.

Deterrence is not just a matter of military hardware and presence. It relies on confidence that commitments will be honoured, risks will be borne, and allies will not be treated with disdain. When US policy becomes more transactional and less predictable, that confidence weakens — even if the underlying military power remains formidable.

But what is the alternative to Pax Americana? Washington’s traditional allies each face their own unique strategic circumstances, and their answers will naturally vary.

In Australia, we have largely managed to keep our head down. We have not been the direct target of American tariffs or sovereignty threats like Canada and Europe. Nor have we publicly challenged Washington in the way some others have – most recently in response to Trump’s apparent contempt for allied sacrifice.

Instead, Australia has doubled down on alliance management. This is mostly visible through AUKUS, which is hanging on doggedly despite growing questions about timeframes, costs and long-term sustainability.

AUKUS reflects Canberra’s judgement that remaining deeply embedded in the US strategic system is preferable to standing outside it. But it also exposes the Australian government to charges it is accepting new forms of dependence on future American and British political decisions, industrial capacity we do not control, and timelines that stretch beyond the current strategic decade.

It is a wager on alignment and continuity at a moment when both are uncertain. That reality frames how Australia should respond to Carney’s call.

---

Read more:
The end of ‘Pax Americana’ and start of a ‘post-American’ era doesn’t necessarily mean the world will be less safe

---

Eroding confidence and trust in the US

Throughout the post-war era, Australian governments have spoken about the US alliance in warm, expansive terms: shared values, shared history, shared sacrifice. The relationship was framed not only as strategically necessary, but morally reassuring. That language is becoming hard to sustain.

Public confidence in the United States has weakened considerably since Trump took office again and began pushing an “America First” doctrine. In public debate, criticism of American conduct increasingly competes with, and sometimes displaces, concerns about China’s rising power.

For Australia, this creates an uncomfortable dilemma. The US remains the only power with the military reach and technological depth to shape the regional strategic balance and constrain China’s ambitions.

Yet, the political foundations that made reliance on that power relatively predictable — and domestically saleable — are eroding.

Managing that tension is now a core task of Australian statecraft. The appointment of Greg Moriarty as Australia’s next ambassador to Washington is very welcome. He brings not only deep knowledge of our own military requirements and the US system, but something equally important: long experience in the Asia-Pacific region. He knows better than most that the US-Australia alliance cannot be separated from the dynamics of Australia’s neighbourhood.

But a growing challenge for the Australian government he serves will be to persuade the public that China — rather than the United States — is still our primary strategic problem.

If the public conversation shifts from managing China’s rise to managing America’s decline, governments will struggle to explain why uncomfortable investments, risks and trade-offs with the Trump administration are required.

What unchecked Chinese influence would mean

Australia should maintain cautious about Beijing’s regional behaviour, even while strengthening our bilateral economic ties with China.

The issue is not whether China builds roads, stadiums or ports in the Pacific. It is what an overall environment of uncontested Chinese strategic hegemony in the region would mean for Australia.

If China gains a stronger foothold in the Pacific, regional civil society leaders warn their governments would face pressure to align political positions, security choices and domestic rules with Chinese preferences.

For Australia, the consequences would be profound. Our ability to operate militarily, diplomatically and economically in our own region would narrow. Our capacity to support Pacific partners in resisting coercion would weaken. And our freedom to make independent strategic choices would be constrained.

It is important to acknowledge Canberra is not standing still.

The Albanese government has made real progress in strengthening regional partnerships to help buffer the unpredictable US alliance. This includes the new alliance with Papua New Guinea, recently concluded defence cooperation treaty with Indonesia, and the overall intensified, respectful Pacific engagement we have seen in recent years. All of this reflects a more deliberate effort to embed Australia more deeply in its own region.

These steps deepen Australian influence, give regional partners more choices, and reduce the risks associated with over-reliance on any single external power. But they do not remove the underlying strategic dilemma.

The US still plays an important role in our region, albeit with more caveats than Canberra has traditionally acknowledged.

Let’s be clear. The US does not really contribute much to Pacific economic development and never really has. Its regional relevance lies in its strategic and military weight – the ability to deter high-end conflict and complicate China’s calculations.

But capability is not the same as commitment. Uncertainty itself can be truly destabilising.

American power may still shape the regional environment, but it does so unevenly and with greater risk of miscalculation. China does not need to defeat the US to exploit this; it only needs to test thresholds and capitalise on ambiguity.

Put simply, the protection the US offers is less absolute — and far less reassuring — than Australian rhetoric often implies.

The way forward: not abandonment, but adjustment

First, Australian leaders need to speak more plainly about the US alliance in order to maintain public support.

This means no longer trumpeting shared virtue, but being honest about what is actually a conditional, interest-based arrangement with a larger power whose values and priorities do not always align with our own.

Second, Australia must continue to hedge more deliberately. This includes deepening defence cooperation with Japan and India, enhancing strategic partnerships across Southeast Asia, and sustaining Pacific engagement. All of this becomes more important as US certainty declines.

Third, as others have argued, Australia must invest more seriously in its own capabilities — diplomatically, militarily and politically — so our security is not wholly contingent on a single power.

The era of comforting myths is over. The alliance still matters — but it is more fragile and conditional now. Recognising that is the necessary starting point for safeguarding Australian security.

Ian Kemish is a former head of the prime minister’s international division, and has represented Australia as an ambassador in the Asia-Pacific and Europe. Alongside his UQ role, he is a distinguished fellow at the ANU National Security College and an industry fellow at the Griffith Asia Institute.

– ref. Australia needs to get real about Trump’s changing America – https://theconversation.com/australia-needs-to-get-real-about-trumps-changing-america-274424

Evening Report: https://eveningreport.nz/2026/01/30/australia-needs-to-get-real-about-trumps-changing-america-274424/

Source: The Conversation (Au and NZ) – By Kai Riemer, Professor of Information Technology and Organisation, University of Sydney

How do you translate ancient Palmyrene script from a Roman tombstone? How many paired tendons are supported by a specific sesamoid bone in a hummingbird? Can you identify closed syllables in Biblical Hebrew based on the latest scholarship on Tiberian pronunciation traditions?

These are some of the questions in “Humanity’s Last Exam”, a new benchmark introduced in a study published this week in Nature. The collection of 2,500 questions is specifically designed to probe the outer limits of what today’s artificial intelligence (AI) systems cannot do.

The benchmark represents a global collaboration of nearly 1,000 international experts across a range of academic fields. These academics and researchers contributed questions at the frontier of human knowledge. The problems required graduate-level expertise in mathematics, physics, chemistry, biology, computer science and the humanities. Importantly, every question was tested against leading AI models before inclusion. If an AI could not answer it correctly at the time the test was designed, the question was rejected.

This process explains why the initial results looked so different from other benchmarks. While AI chatbots score above 90% on popular tests, when Humanity’s Last Exam was first released in early 2025, leading models struggled badly. GPT-4o managed just 2.7% accuracy. Claude 3.5 Sonnet scored 4.1%. Even OpenAI’s most powerful model, o1, achieved only 8%.

The low scores were the point. The benchmark was constructed to measure what remained beyond AI’s grasp. And while some commentators have suggested that benchmarks like Humanity’s Last Exam chart a path toward artificial general intelligence, or even superintelligence – that is, AI systems capable of performing any task at human or superhuman levels – we believe this is wrong for three reasons.

Benchmarks measure task performance, not intelligence

When a student scores well on the bar exam, we can reasonably predict they’ll make a competent lawyer. That’s because the test was designed to assess whether humans have acquired the knowledge and reasoning skills needed for legal practice – and for humans, that works. The understanding required to pass genuinely transfers to the job.

But AI systems are not humans preparing for careers.

When a large language model scores well on the bar exam, it tells us the model can produce correct-looking answers to legal questions. It doesn’t tell us the model understands law, can counsel a nervous client, or exercise professional judgment in ambiguous situations.

The test measures something real for humans; for AI it measures only performance on the test itself.

Using human ability tests to benchmark AI is common practice, but it’s fundamentally misleading. Assuming a high test score means the machine has become more human-like is a category error, much like concluding that a calculator “understands” mathematics because it can solve equations faster than any person.

Human and machine intelligence are fundamentally different

Humans learn continuously from experience. We have intentions, needs and goals. We live lives, inhabit bodies and experience the world directly. Our intelligence evolved to serve our survival as organisms and our success as social creatures.

But AI systems are very different.

Large language models derive their capabilities from patterns in text during training. But they don’t really learn.

For humans, intelligence comes first and language serves as a tool for communication – intelligence is prelinguistic. But for large language models, language is the intelligence – there’s nothing underneath.

Even the creators of Humanity’s Last Exam acknowledge this limitation:

High accuracy on [Humanity’s Last Exam] would demonstrate expert-level performance on closed-ended, verifiable questions and cutting-edge scientific knowledge, but it would not alone suggest autonomous research capabilities or artificial general intelligence.

Subbarao Kambhampati, professor at Arizona State University and former president of the Association for the Advancement of Artificial Intelligence, puts it more clearly:

Humanity’s essence isn’t captured by a static test but rather by our ability to evolve and tackle previously unimaginable questions.

Developers like leaderboards

There’s another problem. AI developers use benchmarks to optimise their models for leaderboard performance. They’re essentially cramming for the exam. And unlike humans, for whom the learning for the test builds understanding, AI optimisation just means getting better at the specific test.

But it’s working.

Since Humanity’s Last Exam was published online in early 2025, scores have climbed dramatically. Gemini 3 Pro Preview now tops the leaderboard at 38.3% accuracy, followed by GPT-5 at 25.3% and Grok 4 at 24.5%.

Does this improvement mean these models are approaching human intelligence? No. It means they’ve gotten better at the kinds of questions the exam contains. The benchmark has become a target to optimise against.

The industry is recognising this problem.

OpenAI recently introduced a measure called GDPval specifically designed to assess real-world usefulness.

Unlike academic-style benchmarks, GDPval focuses on tasks based on actual work products such as project documents, data analyses and deliverables that exist in professional settings.

What this means for you

If you’re using AI tools in your work or considering adopting them, don’t be swayed by benchmark scores. A model that aces Humanity’s Last Exam might still struggle with the specific tasks you need done.

It’s also worth noting the exam’s questions are heavily skewed toward certain domains. Mathematics alone accounts for 41% of the benchmark, with physics, biology and computer science making up much of the rest. If your work involves writing, communication, project management or customer service, the exam tells you almost nothing about which model might serve you best.

A practical approach is to devise your own tests based on what you actually need AI to do, then evaluate newer models against criteria that matter to you. AI systems are genuinely useful – but any discussion about superintelligence remains science fiction and a distraction from the real work of making these tools relevant to people’s lives.

Kai Riemer is co-author of the annual “Skills Horizon” research project, which identifies key leadership skills (including in AI), based on interviews with global and Australian leaders and executives across various fields. He also educates leaders in AI fluency through Sydney Executive Plus at the University of Sydney.

Sandra Peter is co-author of the annual “Skills Horizon” research project, which identifies key leadership skills (including in AI), based on interviews with global and Australian leaders and executives across various fields. She also educates leaders in AI fluency through Sydney Executive Plus at the University of Sydney.

– ref. AI is failing ‘Humanity’s Last Exam’. So what does that mean for machine intelligence? – https://theconversation.com/ai-is-failing-humanitys-last-exam-so-what-does-that-mean-for-machine-intelligence-274620

Evening Report: https://eveningreport.nz/2026/01/30/ai-is-failing-humanitys-last-exam-so-what-does-that-mean-for-machine-intelligence-274620/