Imagine commissioning an engineer to design a bridge. You pay a substantial fee, confident that the engineer has the qualifications, experience and professional judgement to do the job.
The bridge is built. It collapses. People are killed.
Would we shrug our shoulders and say that engineering is complicated, nobody can predict everything, and perhaps we should learn some lessons for next time?
Of course not.
There would be investigations. Questions about the calculations, materials, assumptions and safety margins. If negligence were established, the engineer could face substantial financial liability, professional sanctions and, in sufficiently serious circumstances, criminal prosecution.
And rightly so.
So why do we appear to apply a different standard to other experts whose advice can affect thousands, even millions, of people?
Consider the latest embarrassment involving Deloitte, one of the world's largest professional consulting firms.
The Western Australian government commissioned a $166,000 feasibility report examining possible university mergers.
The report contained extraordinary financial discrepancies. In one scenario, figures adding up to $659 million were presented as $399 million—a difference of $260 million.
There were also apparently AI-generated references, including one containing ChatGPT metadata.
Deloitte acknowledged using generative AI to assist with aspects of the report. The company has maintained that the errors did not affect the underlying modelling or conclusions.
Perhaps they didn't.
But here's my question: If a consultancy cannot reliably check the arithmetic and references in its own report, why should we trust its more complicated conclusions?
The government is seeking reimbursement. That is a start.
But it raises a much larger question about the responsibilities of professional advisers.
We don't pay experts simply to produce documents. We pay them for their knowledge, judgement, diligence and willingness to stand behind their work.
If the result is wrong because they failed to exercise reasonable professional care, why should their liability end with an apology or a refund?
Then there is Snowy 2.0.
When this enormous pumped-hydro project was announced in 2017, Australians were told it would cost around $2 billion and begin operating in 2021.
The latest assessment, released in October 2026, puts the target cost at $22.2 billion, with a possible $25 billion outcome. Full commissioning has been pushed back to late 2029.
That's more than twelve times the original publicly announced figure.
Twelve times!
Now, I appreciate that Snowy 2.0 is a technically demanding project. Geological conditions are difficult. Labour and material costs change. Unexpected problems arise.
But surely that is precisely why we employ experts.
We don't need highly paid consultants to tell us how much a project would cost if everything went perfectly. We need them to assess what could go wrong, quantify the risks, provide realistic ranges and tell decision-makers when the uncertainties are too great.
The Australian National Audit Office has identified deficiencies in Snowy 2.0's project governance and risk management, with the financial risks increasingly falling on government.
Politicians certainly bear responsibility. They announce projects, approve expenditure and often have powerful incentives to present attractive cost estimates.
But somebody supplies the numbers.
Who produced the original estimates? What assumptions did they make? What warnings were given? Were those warnings communicated honestly to ministers and the public? And did anyone knowingly present an optimistic scenario as a reliable budget?
These questions deserve answers.
A twelvefold cost increase does not, by itself, prove negligence. But it surely justifies an independent examination of the original advice and the professional standards applied.
If the experts did their jobs properly and politicians ignored their warnings, we should know that too.
Accountability should follow the evidence, wherever it leads.
The consequences of questionable expert advice can extend far beyond financial losses.
During COVID, governments relied heavily on epidemiologists, public health officials and other specialists to justify extraordinary restrictions on everyday life.
Businesses closed. Schools shut. Families were separated. People lost jobs, savings and opportunities they could never recover.
Some restrictions undoubtedly helped reduce infections and deaths, particularly when the virus was poorly understood and vaccines were unavailable.
But that cannot be the end of the discussion.
Australia's independent COVID-19 Response Inquiry acknowledged the successes of the early response while identifying serious social, economic and health consequences, including damage to public trust.
The central question is not whether governments should have listened to experts. Of course they should have.
It is whether expert recommendations were subjected to sufficient scrutiny.
Were the benefits of particular restrictions weighed against their wider costs? Were alternative approaches considered? Were assumptions regularly reassessed as evidence changed? Were experts sufficiently candid about uncertainty?
And when restrictions continued long after the original circumstances had changed, who was responsible for demanding a fresh assessment?
There is an important distinction between a decision that proves wrong because circumstances were genuinely unknowable and one that proves wrong because contrary evidence was ignored.
The first is an unavoidable feature of decision-making under uncertainty.
The second may be a failure of professional responsibility.
We need to establish which was which, rather than automatically excusing every mistake as the unavoidable consequence of a crisis.
This brings me to what I believe is the heart of the problem.
People respond to incentives.
Suppose you are a consultant preparing a feasibility study for a government project.
You know the minister wants to announce it. You know your client would prefer a favourable assessment. You know a pessimistic report might make you unpopular and jeopardise future contracts.
And you know that if the project eventually costs three times as much as predicted, you are unlikely to suffer any personal consequences.
What incentive do you have to challenge the assumptions aggressively?
Professional pride? Ethics? Reputation?
All important. But are they enough?
Now imagine a different arrangement.
You know that if your estimates are found to have been prepared negligently, you could lose your professional accreditation, face substantial financial liability and suffer serious reputational damage.
If you knowingly falsify evidence or recklessly conceal critical risks, criminal penalties might follow.
Would you be more careful?
Would you insist on independent verification? Document your assumptions? Disclose worst-case scenarios? Refuse to sign off on figures you couldn't defend?
I suspect you would.
And that is precisely the point.
Accountability is not primarily about punishing mistakes after they happen. It is about creating incentives to prevent avoidable mistakes in the first place.
There is an obvious objection.
Experts cannot guarantee their forecasts. Economies change. Viruses mutate. Geological surprises happen. Even the most diligent professional can be wrong.
Absolutely.
I am not suggesting that every failed prediction should result in a lawsuit, much less a prison sentence.
That would discourage honest advice and encourage defensive, excessively cautious behaviour.
What I am suggesting is that professional responsibility should depend on the quality and integrity of the work, judged against what was reasonably knowable when the advice was given.
An expert who carefully investigates the evidence, discloses uncertainty, tests alternative scenarios and honestly communicates the risks should not be punished merely because events turn out differently.
But an expert who ignores obvious risks, presents speculation as certainty, fails to perform elementary checks or knowingly supplies misleading figures should face consequences proportionate to the misconduct and resulting harm.
We already recognise such distinctions in engineering, medicine, accounting and other professions.
Why should major public-policy and management advice be treated differently?
I would like to see several changes.
First, major government projects should require genuinely independent assessment of costs, benefits and risks before political commitments are made.
Second, every major consultancy report should be personally signed and certified by the senior professionals responsible for its conclusions.
Not merely issued under a corporate logo.
The signatories should identify themselves, their qualifications and their responsibilities, and formally certify that the report has been prepared with appropriate professional care, that material assumptions have been tested, and that significant uncertainties have been disclosed.
Their signatures should carry legal and professional consequences.
If serious negligence is subsequently established, those individuals should not be able to hide behind the corporate entity that employed them.
This would not make them guarantors of every forecast. It would make them personally accountable for the professional integrity of the work they have certified.
Imagine how much more carefully a consultant might examine a report if their own name, reputation, professional standing and financial liability were attached to it.
Third, assumptions, uncertainty ranges and significant dissenting opinions should be documented and, wherever possible, publicly available.
Fourth, professional advisers should face meaningful financial and disciplinary consequences where negligence or serious breaches of professional standards are established. Appropriate professional indemnity insurance should be part of the system, not a substitute for personal responsibility.
And finally, deliberate falsification, fraud or sufficiently reckless misconduct causing serious harm should be capable of attracting criminal sanctions, subject to proper investigation and due process.
Not because an expert's forecast was wrong.
Because the expert failed in a duty that society had every right to expect them to fulfil.
Politicians, of course, must also be accountable for decisions they make, particularly when they ignore professional warnings.
The expert's responsibility is to provide competent, honest advice.
The politician's responsibility is to decide what to do with it.
Neither should be able to hide behind the other.
Today, the rewards of expert advice are often private, while the consequences of defective advice fall on the public.
The consultant collects the fee. The politician gets the announcement. The project blows its budget. The taxpayer pays.
Or expert advice underpins a damaging public policy, and those who suffer its consequences are offered little more than an assurance that lessons have been learned.
That is not good enough.
We cannot eliminate uncertainty or punish every honest mistake. But we can demand professional diligence, transparency and accountability.
Make the experts put their names to their work. Make them responsible for the professional standards behind their conclusions. And when serious negligence is established, make the consequences real.
Because accountability does more than punish failure. It changes the incentives that lead to failure in the first place.
We expect that of the engineer who designs a bridge.
Why should we accept anything less from the experts whose advice can cost billions, destroy livelihoods or restrict our freedoms?
Grappy
Millions of people believe they are allergic to one of medicine's most useful antibiotics. New medical guidelines suggest that most of them may be wrong.
Imagine being told as a child that you are allergic to penicillin.
Perhaps you developed a rash while being treated for an infection. Your doctor, understandably cautious, recorded a penicillin allergy in your medical history.
From that day forward, every doctor who treated you avoided prescribing penicillin.
Twenty, forty, even sixty years later, the warning is still there.
But what if you were never allergic in the first place?
According to new Australian medical guidelines, that possibility is far more common than most people realise.
On 20 October, the Australasian Society of Clinical Immunology and Allergy (ASCIA) released updated guidelines for assessing penicillin allergies in adults.
The figures are remarkable.
Approximately 10% of people report being allergic to penicillin, yet fewer than 1% of the population are thought to have a genuine allergy.
More than 90% of people carrying a penicillin-allergy label can tolerate the antibiotic following appropriate medical assessment and testing.
In Australia alone, an estimated 2.8 million people have a reported penicillin allergy.
That suggests an extraordinary number of people may be living with a medical warning that is no longer valid—or perhaps never was.
And this is not just an Australian phenomenon.
There are several explanations.
A child taking penicillin develops a rash. The rash may actually have been caused by the infection rather than the antibiotic.
Sometimes an ordinary side effect, such as nausea or diarrhoea, is mistaken for an allergic reaction.
In other cases, a genuine childhood allergy may disappear over time.
But once the allergy is entered into the medical record, it tends to remain there.
Future doctors understandably take the warning seriously. Few have the time or opportunity to investigate an event that may have occurred decades earlier.
The result is a medical label that can become virtually permanent, even when the evidence supporting it is questionable.
Penicillin and related antibiotics remain among medicine's most valuable treatments for bacterial infections.
When doctors believe a patient is allergic, they may prescribe an alternative antibiotic.
Sometimes that alternative is entirely appropriate. But it may also be less effective for the particular infection, more expensive or more likely to cause unwanted side effects.
Some alternatives are broad-spectrum antibiotics, which affect a wider range of bacteria than necessary.
Their unnecessary use can contribute to antibiotic resistance—one of modern medicine's most serious challenges.
An incorrect allergy label can therefore have consequences extending well beyond the individual patient.
It can influence treatment outcomes, healthcare costs and the effectiveness of antibiotics for everyone.
There is an important qualification.
Genuine penicillin allergies exist, and some reactions can be life-threatening.
Nobody should simply disregard an allergy recorded in their medical history or experiment with taking penicillin.
The new guidelines instead encourage properly supervised reassessment.
A doctor can review the original reaction, determine the level of risk and decide whether further testing is appropriate. Some low-risk patients may be assessed in general practice, while others require specialist investigation.
The objective is not to remove allergy warnings indiscriminately. It is to distinguish those that are justified from those that are not.
The process even has a name: penicillin allergy de-labelling.
What interests me most about this story is the broader lesson.
Modern medicine rightly places enormous importance on maintaining accurate patient records.
But there is a difference between recording a diagnosis and establishing that it remains correct.
A warning entered decades ago can be repeated so often that nobody thinks to question it.
Every subsequent doctor acts responsibly by respecting the record. Yet the combined result may be unnecessarily restricted treatment choices for millions of patients.
Nobody needs to have made a serious mistake for the system to produce an undesirable outcome.
Perhaps we should pay as much attention to reviewing old diagnoses as we do to recording new ones.
If you have been told you are allergic to penicillin, particularly following a childhood reaction, the new guidelines suggest a simple question worth discussing with your doctor:
Am I genuinely allergic, or have I simply been carrying the label all these years?
The answer could make a significant difference the next time you need antibiotics.
And it leaves us with a wider question.
How many other things in our medical records do we continue to accept as facts, simply because nobody has thought to check them?
Patient information: ASCIA — Penicillin Allergy: Frequently Asked Questions
An ancient skeleton, a mysterious wound and a discovery that challenges what we thought we knew about our closest extinct relatives.
Some 75,000 years ago, in a cave in what is now northern Iraq, a Neanderthal suffered a terrible facial injury.
Somehow, this individual survived. The injury healed, suggesting a recovery that may have required help from others.
But there was another injury. A small, circular hole in the shoulder blade, just 15 millimetres across, that researchers believe may have been caused by a wooden spear.
Was this Neanderthal the victim of violence? A hunting accident? We will probably never know.
Yet the damaged bones, recently examined by researchers studying remains from Shanidar Cave, tell a story that challenges our traditional image of Neanderthals.
For generations, Neanderthals were portrayed as primitive, brutish creatures: intellectually inferior cousins of modern humans, destined to disappear when the more sophisticated Homo sapiens arrived.
That picture has been steadily changing.
Archaeological discoveries suggest that Neanderthals made sophisticated tools, controlled fire, hunted cooperatively and lived in complex social groups. There is also evidence that they sometimes cared for injured or disabled companions.
At Shanidar Cave, several Neanderthal individuals were found close together, raising the possibility that the dead were deliberately placed there.
The evidence is not conclusive. Surviving a serious injury does not prove that someone provided nursing care, and the apparent spear wound does not establish murder.
Nevertheless, a picture is emerging of creatures far more complex than the lumbering brutes of popular imagination.
They suffered injuries, survived hardships, formed relationships and perhaps mourned their dead.
Does that sound so different from us?
Neanderthals were not alone.
While they occupied much of Europe and western Asia, another human population was developing in Africa: Homo sapiens, our direct ancestors.
Eventually, the two populations encountered one another.
Were they enemies? Did they compete for territory and food? Did they sometimes cooperate?
We don't know. Their encounters may well have involved all these possibilities.
But one thing is certain. They interbred.
Modern genetic research has established that Neanderthals and Homo sapiens had children together, and those children had descendants.
The implications are extraordinary.
Most people with ancestry outside Africa carry approximately 1–2% Neanderthal DNA. Smaller amounts also occur in some populations with predominantly African ancestry.
Different people carry different fragments of that inheritance, which means that collectively, humanity retains a substantial portion of the Neanderthal genetic legacy.
Some inherited variants influence our immune systems and other biological characteristics.
So Neanderthals did not simply disappear without a trace. Their genes became part of our own evolutionary story.
This does not mean that the individual discovered at Shanidar Cave was one of our ancestors. There is no evidence of that.
Nor do we know whether the companions who may have helped that individual survive were Neanderthals or Homo sapiens. The evidence offers no basis for identifying them.
But the broader discovery is unmistakable: the boundary between these ancient human populations was not absolute.
Neanderthals disappeared as a distinct population roughly 40,000 years ago. Why remains debated. Climate change, competition, small populations and interbreeding may all have played a part.
For a long time, we imagined their disappearance as the triumph of modern humanity over a lesser species.
The reality appears considerably more complicated.
Seventy-five thousand years separate us from the individual whose bones were discovered in Shanidar Cave.
We know nothing of that person's name, relationships, fears or hopes. We cannot know whether the final injury resulted from an accident or an act of violence.
Yet those bones suggest a life marked by injury, survival and perhaps the kindness of others.
And modern genetics tells us something even more remarkable.
Neanderthals were not merely another kind of human who vanished into prehistory. Part of them lives on in us.
Perhaps the greatest surprise is not how different they were.
It is how much we have in common.
