
We give airline pilots aptitude tests, psychometric assessments, medical examinations, simulator checks, security vetting, and pages of paperwork. Then we close the cockpit door, put perhaps 300 people behind them, and effectively say “off you go, we trust you.” That trust is fundamental to commercial aviation. Most passengers will never meet the people sitting behind the locked door at the front.
They don’t know their backgrounds, their personalities, or what sort of day they’re having. They simply trust that somebody, somewhere, has made absolutely certain those pilots should be there. During my career, I’ve personally flown over a million trusting strangers safely to their destination. Passengers aren’t simply trusting the pilots: they’re trusting the system that selected them.
A Hot Topic After The flydubai Incident
The recent flydubai flight FZ1073 incident is therefore deeply uncomfortable. According to UAE authorities, on September 30, 2026, a flydubai co-pilot attacked the Captain and attempted to take control of the aeroplane during a flight from Dubai towards Tel Aviv. The circumstances, motive, and background of the individual remain under investigation. I will never speculate about anything investigators have yet to establish.
However, in aviation, we’re entitled to ask a broader question: how well do we really know the people that have our lives in their hands? My argument is simple: we should keep the aptitude tests, keep the psychology, keep the simulator assessments and background checks. However, we should ensure experienced operational captains remain at the heart of pilot selection. This is because professional judgement cannot be measured entirely on a spreadsheet or by an algorithm.
We Made Aircraft Safer
In the early decades of commercial aviation, a large part of the safety battle involved the aircraft itself. Engines failed more frequently. Systems were less reliable. Navigation and weather forecasting were less sophisticated. Structural and mechanical failures played a much greater role in accidents than they do today.
Over successive generations, aviation became extraordinarily good at engineering those risks out. Engines became remarkably reliable. Aircraft gained multiple layers of redundancy. Navigation improved. Maintenance became more sophisticated. Flight-data monitoring helped us identify problems before they became accidents.
The machines became safer, and accident rates fell dramatically. As that happened, aviation’s attention increasingly turned towards the human being. Crew Resource Management, Threat and Error Management, human-factors training, fatigue management and Safety Management Systems all developed.
They did so because we recognised that people (pilots, engineers, controllers, managers and organisations) had become an increasingly important part of the remaining risk. We became very good at understanding how decent, competent professionals can make mistakes. But perhaps there is another human-factors question that deserves more attention: who exactly are we allowing into the cockpit in the first place?
September 11 & The Germanwings Tragedy Changed Aviation Forever
September 11, 2001 demonstrated with horrific clarity how lethal a commercial airliner could become when control of it fell into the wrong hands. Aviation reacted: cockpit doors were reinforced, access became heavily controlled, security screening and intelligence sharing changed, the flight deck became a fortress, and understandably so. However, Germanwings flight 9525 exposed a terrible paradox: what happens when the threat is already inside?
On March 24, 2015, the First Officer of Germanwings flight 9525 deliberately locked the Captain out of the cockpit and commanded the Airbus A320 into a descent towards the French Alps. All 150 people aboard died. The reinforced cockpit door, designed to prevent unauthorised access, meant the Captain could not regain control of the aircraft. The subsequent investigation found that the First Officer had been suffering from a serious mental-health condition, including psychotic symptoms.
He had also consulted several doctors before the accident. Medical certificates declaring him unfit for work had been issued, but the relevant information did not reach all of the people or organisations who might otherwise have prevented him from flying. The investigation highlighted weaknesses in the interaction between medical confidentiality, self-reporting and aviation medical oversight. That is an enormously difficult area, as the confidentiality between a pilot and their doctor is essential.
Pilots must be able to seek help for anxiety, depression, stress or other mental-health problems without believing that admitting to a difficulty will automatically destroy their career. If we frighten pilots into hiding problems, we make aviation less safe, not more. Airlines should encourage pilots to come forward early, provide confidential support and help them recover and return to flying safely whenever possible.
Mental-health treatment should never carry a stigma, but there must also be a clear route for those rare situations where a genuinely serious condition creates an immediate risk to hundreds of other people. The pilot deserves confidentiality and support, but the passengers also deserve protection.
We Test Almost Everything
Modern pilot selection is remarkably sophisticated. European regulations require psychological assessment before commercial pilots begin line flying. Airlines conduct medical examinations, simulator assessments, aptitude testing, security and criminal record checks, licence verification and reviews of training and employment history. Following the Germanwings tragedy, European aviation strengthened psychological assessment, peer-support programmes and aero-medical safeguards.
All of that is valuable, and none of it should disappear, but there is another form of assessment that is considerably harder to quantify: professional judgement. When I entered commercial flying, getting anywhere near a major airline’s cockpit generally required substantial experience. Several thousand hours were often expected, and those hours did far more than teach you how to fly: they created a track record.
You had worked for people, flown with people, made decisions, made mistakes, been under pressure, and probably frightened yourself once or twice. However, most importantly, you had developed a reputation. Aviation was a relatively small world: if somebody was brilliant, people knew, and if somebody was difficult, people also knew.
If somebody demonstrated poor judgement, a dangerous attitude or behaviour that concerned other pilots, people often knew that too. Your CV told an airline what you had done, and your reputation told them who you were. Today a pilot can go directly from flying school to the right seat of an airliner with hundreds of hours, not thousands, with no employment history, just a licence and exam results.
The Captain At The End Of The Table
I remember airline interviews where a very experienced Senior Captain sat on the interview board. He wasn’t there to check my technical knowledge or administer a personality test: he talked to me. He asked why did you leave your last job, tell me about a captain you disagreed with, what’s the worst decision you’ve made in an aircraft, have you ever broken a rule, and what would your previous Chief Pilot say about you?
The questions could appear conversational, but the captain was doing something extremely sophisticated: he was forming a judgement. Could I sit beside this person for twelve hours? Will they challenge me if I’m wrong? Can they take criticism? Will they admit a mistake? What happens when their ego gets bruised? Ultimately, would I trust this person when things start going wrong?
An experienced captain may have spent tens of thousands of hours observing people operating when they are tired, stressed, irritated, frightened, and under pressure. That experience is invaluable. Malcolm Gladwell explored something similar in his book Blink, describing what he calls ‘thin-slicing,’ the ability of experienced people to recognise patterns and form surprisingly accurate judgements from relatively small amounts of information.
An experienced airline commander may have spent decades sitting beside hundreds, perhaps thousands, of different pilots and personalities. Over time you develop an instinct for how people communicate, how they respond when challenged, whether they listen, whether ego gets in the way and whether something simply doesn’t feel right.
That instinct should never replace proper psychological assessment, evidence or structured selection, but neither should we dismiss the accumulated judgment of somebody who has spent 20 or 30 years observing pilots at close quarters. Sometimes an experienced captain sitting across an interview table can sense something that a test can’t measure.
Background Checks Should Mean Something
Background screening also needs to be more than a box-ticking exercise. Previous employment should be examined properly and thoroughly, and references should be meaningful. Where legally possible, genuine safety-relevant concerns need a route by which they can follow somebody through the industry.
Screening should also not necessarily end on the day somebody is hired. After all, people change, circumstances change, behaviour changes, and sometimes warning signs appear later. Aviation is extremely good at recurrently checking whether I can handle an engine failure or rejected takeoff.
However, perhaps we should be equally interested in ensuring that the human being operating the aircraft remains somebody we can trust with that responsibility. Again, that doesn’t mean treating mental illness as misconduct. Someone responsibly seeking help for depression, anxiety or stress should be supported, not punished.
The objective is not to employ pilots who have never experienced difficulties, as such people probably don’t exist. Rather, the objective is to identify genuinely concerning behaviour, judgement or conditions that might create an unacceptable safety or security risk while providing proper support for pilots who ask for help.
Perhaps We’ve Become Too Clever
Modern aviation loves systems. SMS, CRM, TEM, FRMS: we have acronyms for virtually everything, so I’ve got another one. CSMS, or Common Sense Management System, whereby the first principle could be remarkably simple: when selecting somebody who may one day have the lives of several hundred people behind them, don’t merely ask if this person can pass our selection process.
Rather, we should ask who this person is, talk to people who have flown with them, look properly at their professional history, and allow genuinely safety-relevant information to be shared through lawful and responsible channels. We should also keep appropriate medical and security oversight alive throughout a career, and put experienced pilots into the pilot selection process.
Technology can tell us an enormous amount about a candidate, but psychology can tell us more. After 45 years in aviation, I still believe there’s enormous value in looking another pilot in the eye, talking to them, and allowing somebody with decades of operational experience to ask one simple question: would I trust this person sitting beside me at 38,000 feet if things went horribly wrong?
The Bottom Line
September 11 taught us what can happen when the wrong people gain control of an airliner. The Germanwings tragedy taught us that the threat does not necessarily have to break through the cockpit door, and more recent incidents remind us that however sophisticated our aircraft become, the human beings operating them remain fundamental to safety. Passengers should never have to wonder who is behind the cockpit door.
They should be able to board an aircraft, fasten their seatbelt, and trust that the people at the controls have been selected not only for their qualifications, but for their judgement, temperament and character. After all, flying an airliner is the easy part to measure, but trust is harder. As such, when hundreds of lives depend on the person sitting at the controls, we owe it to every passenger to make sure we are measuring both.







