The secret world of political lobbying has been thrown into the spotlight by ABC Four Corners this week. The program examined the loopholes in lobbying regulations in the context of the Albanese government’s recent gambling reforms.

Today, new research from the Australian Democracy Network uncovers core flaws in Australia’s lobbying rules.

The current system allows powerful people, organisations and corporations to attempt to influence government decision-making covertly. It makes it almost impossible for the public to know who is shaping important policy and why.

Lobbying in Australia

Lobbying is any activity carried out to influence a government or institution’s policies and decisions in favour of a specific cause or outcome.

In Australia, anyone can lobby governments and has a right to represent their views. This is crucial to a healthy democracy.

But Australians also have a right to know who is trying to influence their government.

Not everyone has the same opportunities to access and influence government. Former ministers and senior advisors that become lobbyists can offer government experience and the professional networks that accompany it.

Under the current system, commercial interests can pay for that political access and influence in ways ordinary people cannot. Because there are no records of lobbyists’ meetings with ministers and members of parliament, we have no way of knowing the purpose or outcomes of these meetings, what was discussed, or what tactics might be being used to persuade government ministers to act in line with a lobbyists’s vested interest.

Without safeguards, vested interests can influence government without the public knowing about it.

To protect against undue influence, Australia has a lobbying code of conduct to regulate lobbyists. But this is not a law: it’s not legally binding.

Yet under Australian rules, only one type of lobbyist is regulated: third-party lobbyists. This means a person or company conducting lobbying activities on behalf of a third-party client.

This only covers consultants who provide lobbying as a fee-for-service arrangement. It does not cover “in-house” lobbyists — such as “government relations” staff employed directly by major companies, industry associations and not-for-profits.

The code also prohibits certain former government representatives from lobbying for set periods after leaving office. This applies to:

• ministers

• assistant ministers

• ministerial advisers

• senior public servants

• high-ranking members of the Australian Defence Force.

Only these categories must be identified as former government representatives on the lobbyist register.

Integrity experts estimate that because of these narrow definitions, as many as 80% of lobbyists may fall outside the current rules.

Research published last year by Transparency International Australia found the federal rules are weaker than every state jurisdiction.

This gap is set to widen as four states and territories have reviews underway to further strengthen their own lobbying rules.

What our research found

My research explores how business interests influence public health policies. I helped advise and review the new report from the Australian Democracy Network that reveals major shortcomings in the design and implementation of our federal lobbyist register.

The network’s research found more than half (58%) of ASX100 companies were not listed as clients on the federal lobbying register. This includes some of Australia’s largest companies, such as the Commonwealth Bank, Santos and Qantas. All these companies have substantial commercial interests in policy outcomes.

One explanation is that these companies may not lobby the government.

But a more likely explanation is these companies have their own in-house lobbyists (such as a “government relations” team). The lobbying code of conduct does not recognise these teams as lobbyists.

There are 714 registered lobbyists at the federal level, compared to only 226 members of parliament. Using publicly available records and a broader definition of former government representative — any prior role in state or federal government — the Australian Democracy Network’s analysis found almost two thirds of these lobbyists have prior government experience.

While 295 disclosed this, a further 164 had government experience they either did not disclose or were not required to disclose (such as previously being a member of parliament, or experience in state politics).

4 key reforms

These findings matter because they show how loopholes in our lobbying regulations can camouflage political access and influence.

Compared to international lobbying registers, of which some require companies to disclose their total lobbying spending and have broader definitions of “lobbyist”, Australia’s federal register provides limited information.

Australians deserve to know who is shaping the government decisions that ultimately influence their lives. Right now, it is exceptionally time-consuming (and sometimes impossible) to find out who is trying to influence government policy.

To fix these loopholes, Australia needs a federal lobbying act with universal rules applied to all lobbyists. The act should include:

1. an expanded definition of lobbyist to capture in-house “government relations” staff working for companies and industry organisations

2. a five-year ban on former ministers and advisers becoming lobbyists to protect public interest decision-making from undue influence

3. a requirement for the meeting diaries for ministers, MPs, senior advisers and lobbyists to be published so the public can see who is meeting with decision-makers

4. and the creation of an independent regulatory body with the power to enforce the rules and penalise non-compliance.

Earlier this year, the Speaker of the House and President of the Senate moved to tighten parliamentary access rules for lobbyists. Among other changes, lobbyists with privileged access passes (passes that allow them to wander Parliament House without an escort) will now be listed on a public register.

It’s a welcome first step, but the rest of Australia’s lobbying rules urgently need to catch up.

The author would like to acknowledge Christian Slattery from the Australian Democracy Network for his work on the report that underpins this article.

Jennifer Lacy-Nichols, Senior Research Fellow in Commercial Determinants of Health, The University of Melbourne

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Tech giant Meta will pay up to roughly US$18 billion, plus make major design changes to Facebook and Instagram, as part of a settlement it has agreed to in its most recent lawsuit.

Filed in 2023 by a coalition of 29 US state attorneys-general, the case alleged that Instagram and Facebook were engineered to keep teenagers hooked, that Meta misled the public about the harm, and that it collected data from children aged under 13 without parental consent.

The trial began just over a week ago. But instead of battling it out in court for four more weeks, the parties agreed to settle.

Even though Meta folded in this case, the settlement doesn’t amount to an admission of wrongdoing, liability or violation of any laws. In a statement, Meta said:

Ensuring teens have a safe and productive experience on our platforms is an absolute imperative for Meta. We want to get this right for parents and teens, and that’s why we partnered with state attorneys general to set a new industry standard.

Why settle now?

This year has been brutal for Meta in court.

In March, a Los Angeles jury found Meta and YouTube liable for harming a young woman who joined Instagram at the age of nine.

The same week, a New Mexico jury found Meta had endangered children, ordering it to pay $567 million to address the youth mental health crisis. In August, a judge lifted that bill to $942 million.

In this most recent case, Meta said the states were seeking as much as $1.4 trillion in penalties. The states denied this, saying the maximum penalties could reach roughly $200 billion.

By settling now, Meta buys certainty and a cap on its public exposure. The states get money and product changes that no legislature has managed to pass.

The money has two layers. Meta will pay roughly $11.7 billion to 47 US states, plus the District of Colombia and three US territories, in ten annual instalments, as well as $1 billion to Texas as part of a separate legal agreement. It will pay a further US$5.3 billion only if TikTok, YouTube and Snap sign up to the same rules and pay their own share.

How Instagram and Facebook will actually change

For “Teen Users” – that is, anyone Meta believes is aged 13–17 – in the participating jurisdictions, the agreement will establish a night mode which blocks access to Facebook and Instagram from midnight to 6am by default, and silences notifications from 10pm to 7am.

A two-hour daily cap applies across both apps combined, although messaging and longform videos that are at least 22 minutes long will not count. Essentially, this draws a line between something that resembles watching a TV show and something that resembles scrolling.

A teen with three Instagram accounts does not get six hours. Meta must link accounts using device IDs and phone numbers.

During school hours (8am to 3pm on weekdays), notifications are muted. Forced breaks appear at 60 and 90 minutes of daily use, with a nudge after 15 minutes of continuous scrolling. Only a parent linked through Meta’s supervision tools can loosen any of these defaults.

The agreement then turns to design. “Like” counts are hidden by default, and “cosmetic procedure filters” are switched off. Filters that reshape your face in ways only surgery could are out. Puppy ears and cat faces are fine. Teens must be offered a non-personalised feed (posts in time order rather than chosen by an algorithm) within ten days of joining and every 90 days after.

Finally, the agreement deals with who is on the platform and what happens to them there. Every user must be age-checked within a year, using tools tested against the international standard.

No more than 10% of 16- and 17-year-olds, and 3% of 13–15-year-olds, may be wrongly classified as adults. Failing to meet these service levels, as determined by an independent auditor, is a breach of the judgement and risks contempt proceedings.

Reports of harmful content must be answered within six hours in 90% of cases. Parents are notified the first time their teen messages an adult, and if the teen repeatedly searches for suicide, self-harm or eating disorder terms.

The auditor will report for five years, and Meta is barred from making misleading claims about its safety features.

Why this matters beyond the US

So far, other countries have taken a different approach to tackling big tech.

Many countries have been looking to Australia, where social media platforms such as Facebook and Instagram are required to take “reasonable steps” to stop under-16s from having accounts.

The new settlement in the US instead accepts 13-year-olds can use social media and regulates the design of the platforms.

Where these two approaches overlap is age assurance – working out how old someone is without asking for a passport. Australian law requires “reasonable steps”; the new settlement in the US sets an error threshold of 10% and 3% for the two respective age categories, tested annually by an outside expert.

The US measures may in turn feed back to Australia, where error thresholds might be incorporated into assessments of age restrictions and the promised response times to reports of harm could influence law reform.

A big question is whether the changes agreed as part of this settlement will spread to other platforms.

The settlement is designed to make this happen: Meta pays the final $5.3 billion only once “industry-wide adoption” occurs, which the agreement defines as Snap, TikTok and YouTube being bound by equivalent rules through settlement, legislation or audited voluntary compliance. That is the same domino logic as the 1998 settlement between 52 US state and territory attorneys-general and the country’s four largest tobacco companies.

If this happens, the night block will widen to 10pm to 7am and the daily cap would drop to one hour per app.

Meta has published an open letter urging TikTok and YouTube to join “right away”, with chief legal officer C.J. Mahoney arguing that “teens move fluidly across dozens of apps”.

Whether TikTok and YouTube can resist the rules that Meta has accepted is the question for the next 12 months.

Rob Nicholls, Senior Research Associate in Media and Communications, University of Sydney

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Some private schools have begun asking parents to sign a code of conduct to try and stop abusive behaviour towards teachers.

In media reports, some principals have called out parents for a sense of entitlement, bullying of staff and refusing to admit when their child has misbehaved.

Public schools are also taking new steps to try and curtail abusive parents.

Victoria and South Australia have introduced laws to give principals powers to ban abusive parents from school grounds. This includes banning parents who are abusive in person, over email and phone or on social media. New South Wales is also considering similar laws.

Why have such strong measures become necessary?

How serious is the problem?

Parents need to be able to question school decisions, raise concerns and advocate for their children.

These are all important parts of meaningful parent engagement with education. But research clearly shows abusive behaviour towards teachers has been growing.

For 15 years, Australian survey research has tracked school leaders’ experiences – this includes principals, deputy principals and other senior leaders. This has found parents and caregivers are a significant and growing source of threatening and abusive behaviour.

In 2025, they were identified as a source in almost two thirds of reported threats of violence – in 2011, this figure was less than 25%. In 2025, nearly nine in ten cases of cyberbullying of school principals came from parents and caregivers.

The impact is not confined to overt threats. Research tells us school leaders also manage persistent complaints, aggressive communication, and expectations of constant availability from parents.

Teachers also report similar experiences, including excessive communication demands, verbal abuse and threatening behaviour.

Why is this happening?

Over recent decades, expectations of what schools can and should provide for students and families have expanded. Greater school choice and accountability have also changed the relationship between families and schools, with increased expectations around access, responsiveness and influence over educational decisions.

Research has identified a more consumer-oriented relationship with schools. For school leaders, this can mean expectations of immediate responses, challenges to professional decisions, demands for individualised outcomes, and complaints being escalated when families are dissatisfied with a school’s response.

Digital communication has made it easier for families and schools to communicate. It has also changed the speed, reach and consequences of these interactions.

For example, concerns can quickly move into parent WhatsApp groups, social media, or other community networks. So disagreements can be amplified well beyond the original interaction.

Laws can protect staff, but …

School staff have the same right to a safe workplace as anyone else.

But laws or codes of conduct only establish boundaries, restrict access and provide recourse when behaviour becomes unsafe. They cannot create respectful behaviour or rebuild trust once a relationship has deteriorated.

This is why legislation should be understood as protection, rather than prevention. If a principal needs to stop a parent entering school grounds or contacting staff, the relationship is already in serious difficulty.

So what can schools and governments do?

No single intervention can prevent conflict between parents and schools.

Clear expectations around communication and staff availability can establish boundaries early. This is where parent codes of conduct or behaviour agreements may be useful, by making expectations about respectful communication and the consequences of unacceptable behaviour explicit.

But there is limited evidence that contracts themselves can prevent or resolve conflict once relationships have seriously deteriorated.

School leaders and teachers also need better preparation to manage difficult conversations with parents, recognise when conflict is escalating, and know when to seek further support. They also need organisational backing, so they are not left managing a situation alone until it becomes untenable. This includes clear escalation process and senior or system-level support when interactions become persistent, unreasonable or abusive.

There will, however, be circumstances when these approaches are no longer enough. When behaviour becomes threatening, intimidating or abusive, the priority must shift from repairing the relationship to protecting the people involved.

What now?

Schools need to remain places where families can engage, question and disagree, while also being a workplace where staff are safe. Neither should come at the expense of the other.

A law can prohibit abusive behaviour, but it cannot produce respectful behaviour. It can protect staff when a relationship has broken down, but it cannot rebuild the trust that has been lost.

That work has to begin much earlier, with clear expectations about communication, complaints and acceptable behaviour established before conflict occurs. Education systems also need to be prepared to intervene when relationships begin to deteriorate.

Carolyn Wade, Lecturer, School of Education, Griffith University and Paul Kidson, Associate Professor in Educational Leadership, Australian Catholic University

This article is republished from The Conversation under a Creative Commons license. Read the original article.

If you’re aged over 70, or over 50 with underlying health conditions, and have tested positive for COVID, you might have been offered free or subsidised antivirals.

Earlier in the pandemic, these medicines were an important way to reduce the chance of people becoming severely ill with COVID, needing to be hospitalised, or dying.

But our new research suggests that with high levels of vaccination and immunity from previous infections, COVID antivirals are no longer working as well.

From no treatment to an effective one

When SARS-CoV-2, the virus that causes COVID, first emerged in late 2019, no treatment was available. Existing influenza antivirals, which had been stockpiled by governments, were ineffective and couldn’t be used.

Scientists rapidly tried to figure out which existing or new drugs could be used. In 2021, two industry-sponsored clinical trials reported promising results.

One was for nirmatrelvir-ritonavir (Paxlovid), which reported an 89% reduction in severe illness. The other was a trial of molnupiravir (Lagevrio), which reported a 30% reduction.

Paxlovid and Lagevrio work by stopping SARS-CoV-2 from replicating. So they need to be taken early in the course of illness – within five days – to be effective.

Both drugs became available in Australia and were listed on the Pharmaceutical Benefits Scheme (PBS) in early and mid-2022. This ensured people most at risk of severe illness had access to drugs at low or no cost.

Paxlovid and Lagevrio are expensive. A course of treatment cost the government around A$1,100 from 2022 to 2025, much more than the treatment for influenza, oseltamivir (Tamiflu), which is only around $40.

Between mid-2022 and mid-2025 the Australian government spent more than $2 billion on Paxlovid and Lagevrio.

What’s different now?

Most Australians have now been vaccinated against COVID and infected with SARS-CoV-2 more than once. This results in hybrid immunity and reduces the likelihood and severity of future infections.

The virus itself has also changed, with the original variants replaced by new ones that appear to be less severe than early in the pandemic.

These changes prompted us to evaluate the more recent evidence.

What did we study and find?

First, we examined randomised controlled trials – the gold standard of evidence. We found eight trials of Paxlovid and Lagevrio, but most were done before people were vaccinated.

When we looked at the two randomised control trials in vaccinated people, neither reported a reduction in severe outcomes after taking either Paxlovid or Lagevrio.

The Lagevrio trial reported no statistical difference in hospitalisation or death in people who did and didn’t receive treatment.

The other was a trial of people in hospital who were treated with Paxlovid. It found no difference in the people treated and not treated.

There were some nuances to these findings. Randomised controlled trials are expensive and resource-intensive. As COVID went from being the leading cause of death worldwide in 2021 to the 20th in 2023, it became more difficult to justify the costs and recruit enough participants, so some trials were stopped early.

However, two randomised controlled trials of Paxlovid have been published since our work, and neither reported a reduction in hospitalisation or death.

In phase two of our research, we looked at the 35 observational studies, where researchers can include many more people but can’t control who gets the treatment. We pooled the results from the studies to estimate the effectiveness for each drug against hospitalisation and death.

This meta-analysis found Paxlovid, but not Lagevrio, reduced hospitalisation and death.

Paxlovid was associated with a 40% reduction in hospitalisation and a 67% reduction in death.

Lagevrio wasn’t associated with a significant reduction in hospitalisation and the results were mixed against death.

Because the treatment wasn’t randomised, there may be differences between people who received treatment and those who didn’t. People accessing treatment, for example, may have greater health literacy or access to other health services.

While some studies use methods to adjust for differences in people who do and don’t receive treatment, it’s usually not possible to adjust for every difference.

COVID is still a concern for at-risk groups

SARS-CoV-2, like influenza and respiratory syncytial virus (RSV), remains a common cause of acute respiratory illness in Australia. In 2025, there were:

• 186,000 cases of COVID reported
• 178,000 cases of RSV
• 503,000 cases of influenza.

But the number of COVID, RSV and influenza cases reported is well below the true number, as most people don’t get tested.

COVID is still claiming lives. There were 2,227 deaths involving COVID in 2025. This is more than the 1,784 deaths involving influenza. Three quarters of deaths were among people aged 80 and over.

The rates of COVID-related hospitalisations so far this year have been lower than last year, with 839 admissions at “sentinel hospitals” (those chosen to monitor disease trends) in the first half of this year. This compares to 1,740 in the same period last year.

What does this mean?

COVID can still cause severe illness, hospitalisation and death, particularly in older people.

If you test positive for COVID, talk to your GP about whether antivirals may be appropriate for you.

Our work shows there may be some benefit from Paxlovid. But as the number of cases and the risk of severe illness continues to fall, this benefit is becoming very small.

Given how little evidence there is for Lagevrio, its role in treating COVID should be carefully considered.

Harms of antivirals can include side effects – such as taste changes, nausea and vomiting – and allergic reactions.

Paxlovid can also interact with other drugs, causing serious reactions. It should be avoided, or the dose adjusted, in people with severe liver or kidney disease because it can make these conditions worse.

Australia’s independent Pharmaceutical Benefits Advisory Committee (PBAC) is set to review COVID antivirals later in 2026. It will assess the costs and benefits of Paxlovid and Lagevrio and it may change its recommendations about who can access the drugs on the PBS.

Our findings support this reassessment.

Laura Edwards, PhD Candidate, Vaccines and Antiviral Effectiveness, UNSW ; Allen Cheng, Professor of Infectious Diseases, Monash University; Bette Liu, Associate Professor and NHMRC Career Development Fellow, School of Public Health and Community Medicine, UNSW , and James Wood, Head of School of Population Health and Professor, Infectious Diseases Dynamics and Interventions, UNSW

This article is republished from The Conversation under a Creative Commons license. Read the original article.

It can take years for someone to be diagnosed with endometriosis. This can mean ongoing pain, repeated appointments and uncertainty, and delayed treatment.

However, two new tests are being trialled in the United Kingdom. One is a saliva test; the other measures electrical activity in the gut. Over the next three years, these non-invasive tests will be used in general practice patients to see if they can help diagnose endometriosis much sooner.

The promise of these and other non-invasive tests to speed up diagnosis is compelling. But the evidence is not there yet for their widespread use.

Why is endometriosis so difficult to diagnose?

Endometriosis is a condition where tissue like the lining of the uterus is found outside the uterus. It commonly causes severe period and pelvic pain, pain during intercourse, bowel and bladder problems, and difficulty becoming pregnant.

Diagnosis involves assessing symptoms, a physical examination, imaging and sometimes surgery. Ultrasound and MRI can diagnose endometriosis when performed by a health professional with specific extra training.

But a normal scan does not rule out endometriosis. So laparoscopy, surgery using a small camera inserted into the abdomen, may still be considered.

Then there are the two non-invasive tests being trialled in the UK but are not available in Australia. What can these offer?

Endotest is a saliva test

Endotest analyses a saliva sample for specific markers called microRNAs – tiny molecules involved in controlling how genes behave. Specific patterns of these microRNAs can indicate whether endometriosis is likely present.

A French study of 971 patients across multiple centres reported the test was highly accurate.

But most participants had endometriosis. That’s unlike typical GP patients presenting with symptoms of pelvic pain, only some of whom will have endometriosis.

Pelvic pain can have bowel, bladder, muscular and other causes. So we don’t know if the test performs as well in the typical mix of patients you’d expect to see in general practice.

EndoSure looks at your gut

EndoSure detects endometriosis by measuring electrical signals in the gut via sensor pads on the abdomen. The test takes 45 minutes and can take place in a GP or specialist clinic.

A study of 154 women reported 95% sensitivity (the ability to correctly detect endometriosis) and 96% specificity (the ability to correctly rule out people who do not have endometriosis).

However, most of the evidence supporting the specificity figure came from controls who did not have surgery to confirm they really did not have endometriosis.

Preliminary findings from a later study compared just 25 women with confirmed endometriosis with 25 controls. Most of the women in the control group had abdominal symptoms but were, again, not surgically confirmed as endometriosis-free.

The study was also not able to establish whether other conditions could interfere with the test. The researchers noted adenomyosis might produce similar signals.

More recent, unpublished findings presented at a scientific meeting found EndoSure detected everyone who had endometriosis. However, its reported specificity was just 5%, compared with around 96% in the earlier study. This means 95% of people without endometriosis still tested positive.

If confirmed in larger studies, that would raise serious questions about its usefulness in general practice, where doctors need to distinguish endometriosis from the many other conditions that can cause similar symptoms.

How about other tests?

Other non-invasive tests for endometriosis are in development, including blood tests. Not all are widely available or have been independently validated.

EndomTest is a blood test available in the United States that combines its results with clinical information. But the US Food and Drug Administration has not approved it for widespread use.

Published results found high specificity for diagnosing endometriosis. But it had much lower sensitivity, meaning it could miss many people who have endometriosis.

Australia’s PromarkerEndo measures proteins in blood. Importantly, testing has involved patients with symptoms but where surgery has found no endometriosis. This is a strong comparison because, when used clinically, the test has to distinguish between patients with similar symptoms who do and don’t actually have the disease. Results have been promising across early-stage disease, although further independent validation is needed.

US-developed DotEndo examines microRNAs in blood. Earlier research found a panel of six-microRNAs could distinguish endometriosis from other gynaecological conditions. A large, multi-centre study is evaluating its use in around 750 patients.

Impressive numbers need context

When we evaluate a diagnostic test, the question is not simply “How accurate is this test?” but “How well does it work in the people who will actually use it?”.

Even so, for a test to be useful, it should change what happens next by raising suspicion, prompting investigation, guiding treatment, or providing an explanation sooner.

So such tests may help answer “Could this be endometriosis?”. But imaging will still be needed to show where and how extensive the disease is. Doctors will still need to assess adhesions – scar-like tissue that can cause pelvic organs to become stuck together.

This mapping guides treatment and, if surgery is considered, its complexity and which specialists may be involved. If such tests help patients access specialist ultrasound earlier, this could be useful.

The same care is needed with a negative result. Unless a test reliably rules out endometriosis in the population being tested, a negative result should not be used to dismiss persistent symptoms. Even without endometriosis, pelvic pain still deserves investigation and care.

Mathew Leonardi, Associate Professor, Gynaecological Surgeon and Sonologist, McMaster University; Adelaide University and Narges Kalani, Clinical Research Assistant, Departments of Obstetrics and Gynecology, and Oncology, McMaster University

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Earlier this week, Queensland Nationals MP Llew O’Brien introduced a so-called “born alive” bill into the House of Representatives.

On the same day, August 17, the Australian Christian Lobby held an event on Parliament House lawns. Against a backdrop of 10,000 knitted baby booties arranged into a cross, Australia’s most prominent federal anti-abortion politicians spoke in support of the bill.

Born alive bills originated in the United States in the early 2000s and the issue has been raised in most Australian decriminalisation debates since Victoria in 2008. O’Brien’s resurrection of this topic reveals much about the ambitions of contemporary Australian opponents of abortion.

What does the bill do?

In Australia, abortion is regulated at the state and territory level. This bill would insert the federal government into the equation.

The Human Rights (Children Born Alive Protection) Bill would require the same medical care for a baby born with signs of life following abortion as for any other baby, a situation that is extremely rare. Any failure to provide such care would be a criminal offence.

Every state and territory in Australia has decriminalised abortion. The bill would reintroduce the threat of criminal penalties for abortion providers with respect to care provided after an abortion. It would likely have a chilling effect on doctors’ willingness to provide abortion services.

This is the third time this bill has been introduced in parliament in the past five years, identical to versions from August 2021 and November 2022.

The 2022 bill went to a Senate inquiry, whose report expressed concerns about its impact on healthcare access, constitutional validity and consistency with human rights.

A human rights bill that isn’t

The bill is framed as human rights legislation, purporting to implement Australia’s international treaty obligations under the Constitution’s external affairs power.

It claims to give effect to rights in the International Covenant on Civil and Political Rights and the Convention on the Rights of the Child.

But it is inconsistent with these treaties and the findings of the United Nations bodies that oversee their implementation.

UN bodies have found abortion restrictions breach human rights, including privacy, protection from cruel, inhuman or degrading treatment, equality and non-discrimination, and the right to life.

Enacting the bill would likely breach Australia’s international obligations. The bill would not meet the requirements of the external affairs power and would likely be constitutionally invalid.

The use of human rights language to support positions inconsistent with human rights norms is no accident.

It mirrors a pattern in anti-abortion action (sometimes called abortion lawfare) in the US and elsewhere.

Scholars argue human rights misappropriations have been “deployed by nationalist, populist, right-wing and authoritarian movements” worldwide.

False narratives

Contemporary born alive bills rest on false narratives.

Supporters vilify abortion providers and frame abortion as barbarism, not health care.

The people who need abortions, meanwhile, are maligned or rendered invisible.

The vast majority of abortions in Australia occur before 20 weeks. About 2% of Australian abortions happen after 20 weeks, typically involving a step to stop the fetal heartbeat first.

Without that step, a fetus may briefly show signs of life. But because their lungs and other organ systems are too immature to sustain life outside the uterus, survival is not possible.

In rare cases, parents skip this step so they can hold their non-viable baby while it dies – a way of grieving, with the baby kept comfortable throughout.

Calling this “born alive” wrongly implies viability.

In 2024, MP Ralph Babet moved an urgency motion to have the Senate “recognise that at least one baby is born alive every seven days following a failed abortion and left to die”.

This claim is extrapolated from a single table in Queensland and Victorian perinatal mortality reports (2010–20), which splits deaths after late abortion into “stillbirth” and “neonatal death”.

Advocates treat the latter as “live birth”, turning a routine statistical classification into evidence for something the data wasn’t designed to measure.

Born alive proponents lean on a handful of cases spanning decades (including from the 1970s and 1990s) and international contexts. Most have little relevance to Australian healthcare practices in the 2020s.

Care that already exists

Born alive bills are opposed by peak medical specialist, nursing and midwifery bodies.

Health professionals must already provide care for any baby born alive after abortion, including palliative or “comfort” care.

Proponents of these bills ignore existing clinical guidelines. Victoria’s guideline, for instance, recommends palliative care for babies born before 22 weeks, and “parental discretion” for resuscitation decisions between 22 and 24 weeks.

So why should parental wishes matter for premature birth, but not for abortion?

The born alive narrative erases the pregnant person, reducing them to a passive vessel. One supporter of the 2022 bill told the Senate the pregnant person is the site where the “child in utero” is “spatiotemporally located”.

Statistics about babies born with no chance of survival showing “signs of life” say nothing about the complex and multifaceted drivers of later abortion.

People seek abortion after 20 weeks when they receive new information, such as a fetal anomaly diagnosis, or when barriers such as reproductive coercion or health system delays make earlier care impossible.

A renewed anti-abortion push

Speakers at the Australian Christian Lobby rally included Nationals Leader Matt Canavan, Liberal Alex Antic, and One Nation’s Barnaby Joyce and Malcolm Roberts.

Roberts frames the born alive bill as a move “to take our country back”.

He recently told an audience of conservative Australian Christians he’ll push One Nation to pursue an abortion ban.

Canavan and Antic are prominent anti-abortion figures within their parties. With Babet, they co-sponsored the 2022 born alive bill.

Joyce and Antic are current collaborators with anti-abortion influencer Joanna Howe.

Howe is prominently aligned with One Nation. She’s actively attempting to reshape the Nationals and Liberals in New South Wales, where anti-abortion legislation will be voted on in September.

What does all this mean?

Without the support of the government, this born alive bill is all but guaranteed to fail.

But its impact lies in giving conservative federal politicians an arena to directly intervene in abortion debates.

It mainstreams health care and legal misinformation, while amplifying stigmatising rhetoric about abortion.

The discussion entrenches the idea that abortion bills can work for partisan messaging and movement-building, advancing the Americanisation of Australian politics.

And it positions abortion as a unifying cause for key right-wing power brokers across the Liberals, Nationals and One Nation.

Prudence Flowers, Senior Lecturer in US History, College of Human Sciences and Culture, Flinders University; Erica Millar, Senior Research Fellow, Social Inquiry, La Trobe University, and Tania Penovic, Associate Professor of Law, Deakin Law School, Deakin University

This article is republished from The Conversation under a Creative Commons license. Read the original article.

For decades, people exercising have been told one simple rule: never let your knees go over your toes. During lunges, squats and many other exercises, well-meaning coaches, trainers and physiotherapists repeated this advice so often it became fitness folklore.

More recently, however, social media has pushed back. Popularised by fitness influencer Ben Patrick – better known as the “Knees Over Toes Guy” – millions have embraced a training style that focuses on building lower limb strength. Among the tenets is the idea that letting your knees travel further forward is a key part of “bullet proofing” your knees.

So who’s right? Like most health and fitness debates, the answer lies somewhere in the middle. It’s not about whether your knees move forward beyond your toes, but what happens when they do.

Knees like to be used

In everyday life, your knees move well beyond your toes all the time. Think about walking downstairs, crouching down to play with kids, and kneeling in the garden.

Imagine doing a squat. As your knees bend further, your thigh muscles work harder to stop your knees from collapsing to the floor. This extra muscle tension increases the force between your kneecap and thigh bone – what we call joint loading.

Now think of exercises like the knees over toes split squat, which look a bit like this.

This movement deliberately allows the knees to travel further forward, increasing this joint loading compared to shallower squats. For some people, this can be painful. For most, however, loading your knees like this is normal and perhaps even a good thing.

Your bones, muscles and articular cartilage (the lining that covers the end of your bones that makes joint movement smooth) are living tissues that need loading to stay healthy. In this context, loading means making them bear weight.

Weight bearing exercise – such as lifting weights – is one of the best ways to keep bones strong (and prevent osteoporosis).

The opposite happens in space, where astronauts rapidly lose bone and muscle because their bodies are no longer exposed to gravity. On Earth, bedrest results in similar changes. Our bodies, including our knees, were designed to be used.

The exercises aren’t actually new

The “knees over toes” approach hasn’t yet been well scientifically tested, so we don’t know whether it performs better than other well-designed strength programs.

However, many of the exercises it promotes, such as split squats and deep squats, have been used for decades. The branding is new, the movement principles are not.

Many of these exercises deliberately bend the knee through a larger range of motion, increasing the work done by your thigh muscles. This may promote greater muscle growth and strength than shallower exercises. Strength training also improves bone health and helps maintain healthy cartilage.

Where caution is needed

The same feature that makes these exercises so useful, their ability to load the knee, is also why they should be progressed gradually.

About one in four people have kneecap pain (sometimes called patellofemoral pain). If that’s you, pain often occurs during activities that place high loads on the knee, such as squatting and walking downstairs.

But pain doesn’t automatically mean damage. In fact, progressively rebuilding your knees’ ability to tolerate heavy loads is one of the most effective treatments we have.

Exercise therapy isn’t simply considered safe – it is the cornerstone of treatment when appropriately prescribed to build back the muscle strength and regain function.

Should you try knees over toes exercises?

The unsatisfying answer is: it depends.

If you’re new to strength training, it’s normal for your muscles to feel worked and even a little sore afterwards.

If you already have knee pain, start with a smaller knee bend, less weight, or fewer repetitions.

For example, one of the signature “knees over toes” exercises is the split squat. Rather than trying to copy the deep versions shown on social media, begin with a comfortable knee bend and gradually increase how far your knee travels forward as your symptoms and strength allow.

Mild discomfort can be acceptable, but pain that steadily builds during exercise, or leaves your knee much worse the next day, usually means you’ve progressed too quickly. That doesn’t necessarily mean you need to stop – try doing a little less before gradually building back up. If you’re unsure, consult a physiotherapist or doctor.

If the exercises are working, you should notice them becoming easier, and everyday activities (such as walking downstairs, getting out of a chair) becoming more comfortable.

Our knees don’t need to stay behind our toes, but they don’t need to be forced past them either.

Whether you choose “knees over toes” exercises or another strengthening program matters less than progressing gradually and consistently.

One exercise from social media is unlikely to be the holy grail for every knee problem or training goal.

The goal is to build a knee that’s strong enough for the life you want to live.

Matt Savage, PhD Candidate and Lecturer in Physiotherapy, La Trobe University and Adam Culvenor, Associate Professor, Principal Research Fellow in Sport and Exercise Medicine, La Trobe University

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Amid all the talk about artificial intelligence (AI) both creating and destroying jobs, a troubling reality flies under the radar.

The tasks machines can’t perform well are often offloaded onto marginalised global workers who are struggling in precarious labour markets. They do ostensibly “automated” work under exploitative conditions.

Data work is an essential part of building and refining AI systems. Before AI models can “learn” anything, human data workers must categorise, label, test and moderate vast volumes of text, images, audio and video, to make the data usable for AI training.

This labour is performed by an expanding global digital workforce that prepares the datasets not only for big tech, but also high-stakes industries such as banking, insurance, healthcare and government agencies, including defence.

To understand the AI workforce, I have been interviewing workers in China and Australia who prepare datasets for AI models. The fieldwork is ongoing, but here’s what they’ve revealed so far.

Inequality is baked in

My interviews with ten people to date show that precarious labour markets and marginalised social status have pushed digitally literate young workers into the data labelling industry.

As one interviewee said:

We do the manual work so that they get the credit for the intelligence.

There’s a lot of inequality across the data labour market, shaped by people’s qualifications and geographic location.

Those with PhD-level or equivalent qualifications and STEM certifications can typically get more specialised tasks. If based in the Global North, such workers tend to be higher-paid, earning A$400–800 per hour depending on the task.

But such specialised and high-paid tasks are rare and difficult to get. Most workers I interviewed perform general tasks, such as repetitively drawing bounding boxes for images used in drones, self-driving cars and automated vending machines, or annotating audio.

These workers normally receive as little as A$6 per day or even less. The pay can’t cover daily expenses, and the long hours leave workers with chronic eye strain and back pain.

Part of the gig economy

Data work is not unlike other poorly regulated jobs in the gig economy.

Workers have no formal contracts and are not employees. They’re classified as “users”, and platforms simply call on them when there are tasks aligning with their expertise and track record.

User agreements exist primarily to protect the companies behind the outsourced work, such as requiring the workers don’t disclose any of the information they see.

This is despite the fact datasets are already anonymised: workers often have no way of knowing which companies they conduct data labelling for. They don’t even know if humans or AI agents assess their completed work. And they have minimal rights to appeal any assessment of their performance.

All interviewees reported getting less work over time as AI advances. What’s left are more difficult and time-consuming tasks. Interviewees expressed little concern about their jobs eventually being replaced by AI, but this apparent indifference stemmed from a pessimistic outlook:

If I don’t make this money, someone else will, and I will be replaced [by AI] eventually anyway.

As one worker noted, what AI actually affects is the working class itself. This working class is expanding as more professionals are pushed into data labelling by the precarity of the current job market.

All work, little pay

How a worker gets paid is determined by the platform. US crowdsourcing platforms generally offer higher-paid tasks and pay workers when they submit the work.

Chinese platforms or companies often pay workers only after their tasks have been assessed and confirmed to meet preset standards. As a result, workers often spend hours completing tasks without receiving any payment.

In addition, workers in China can’t access US platforms; using a VPN to circumvent this risks triggering an account ban.

Companies prefer consistency in their workforce, as turnover is costly; workers require instruction and training before they can begin a task, and further time before they can complete tasks efficiently.

As workers typically get faster the longer they stay in the role, companies want to retain the experienced ones. But many workers leave because the pay is so poor.

To offset this, companies have turned to recruiting more vulnerable groups. One example is collaborating with local government initiatives supporting disabled people. These workers are less likely to quit because the job is often their last resort.

Workers reported they were unable to find other employment or were in the process of searching for full-time positions, due to disability, pregnancy or being recent graduates.

The bigger picture is grim

The AI economy has created jobs. But many of these involve human workers correcting errors and handling tasks too difficult or ambiguous for machines to resolve. This work is often more cognitively and emotionally demanding than what it replaced.

And human workers don’t even know if they’re answering to human managers or AI agents. This weakens their right to bargain.

Data workers are effectively the disposable batteries of the AI economy: drained of every last charge, then discarded once they can no longer power the system that depended on them.

Australia is accelerating the pursuit of an AI-driven economy. The crucial question is not how many jobs are created, but what kind of jobs they are.

The employment gains AI promises may only exist in the short term, and come at the cost of data workers’ life quality and wellbeing. We must establish protections and a long-term plan for this workforce, so we can prevent the harm rather than merely respond to it after the fact.

Fan Yang, Research Fellow at Melbourne Law School, The University of Melbourne

This article is republished from The Conversation under a Creative Commons license. Read the original article.