IATA’s conference opens in Istanbul on Monday under the theme 'Performance Through People' with mental health, peer support and duty of care explicitly on the programme. Hosted by Turkish Airlines, the October 6-8 event brings safety, operations, cabin and crisis-management specialists together as airlines increasingly rely on AI, automation and safety data to manage risk.
The timing is uncomfortable. It follows the flydubai FZ1073 cockpit attack, which UAE prosecutors have called an attempted terrorist act, and renewed scrutiny of how airlines pair crews, monitor fatigue and assess pilots who may be under acute personal or professional strain. No conference policy can explain an active criminal investigation, and no screening system can guarantee prevention. But the event puts a more practical issue in focus: what should an airline do when its safety system sees a human risk before it becomes an emergency?
The most effective approach is not a one-off psychological test at recruitment or an annual box-ticking medical assessment. It is a layered system built around confidential self-reporting, peer support, occupational medicine, fatigue monitoring, operational reporting and clear escalation routes when a credible safety concern arises.
Pilots must be able to disclose depression, anxiety, addiction, trauma, financial stress, sleep problems or acute personal distress without automatically assuming that asking for help ends their career. A system that equates disclosure with punishment encourages concealment. A system that ignores warning signs is equally unsafe.
That balance is likely to define the Istanbul discussion. IATA’s published programme includes a session on integrating mental health, peer support and duty of care into emergency-response planning, as well as workshops on human factors and the way people and systems jointly shape safety performance.
Peer-support programmes are central because fellow pilots may detect changes in behaviour, performance, isolation, conflict or fatigue before a formal clinical process does. But peer supporters cannot be expected to act as covert investigators. Their role should be to provide trusted access to help, encourage voluntary referral and know when a concern crosses the threshold for confidential escalation.
The key safeguard is a defined boundary between confidential support and immediate safety risk. If a pilot presents an imminent risk to self or others, or is unfit to fly, medical and operational intervention must override confidentiality. The rule needs to be understood in advance, not improvised after an event.
IATA has deliberately framed the conference around the human use of technology. Its argument is that data, AI and automation can identify patterns and support decisions, but people still interpret information, make operational judgments and respond to disruption.
That means, for pilots' well-being, airlines can use data carefully to identify organisational risk rather than create a personal-risk score for every aviator. Scheduling systems can detect chronic fatigue exposure, repeated disruptive rosters, inadequate recovery, recurrent short-notice changes or pairing patterns that increase workload. Safety-reporting systems can identify operational stress points. Crew-management teams can then redesign the system causing the strain.
What they should not do is treat AI-generated behavioural inference as a clinical diagnosis or a disciplinary trigger. A predictive model can be wrong, biased or blind to context. If an algorithm marks an employee as 'high risk', the result may be stigma, loss of trust and reduced willingness to use support programmes.
The useful model is therefore human-in-the-loop. Data can flag an operational pattern. Qualified people assess it. Clinicians make medical judgments and pilots retain due process, privacy protections and a route back to flying where medically appropriate.
Traditional screening still matters. Airlines need robust pre-employment checks, aeromedical certification, recurrent medical reviews, security vetting, substance testing where permitted, performance monitoring and mechanisms for colleagues to raise concerns. But screening captures a moment in time. Mental health, fatigue, financial distress, relationship breakdown, grief and burnout can emerge later.
The more difficult work is continuous support. Airlines should ensure that pilots have access to independent counselling, peer networks, confidential reporting and non-punitive pathways to temporary removal from duty. They should also make return-to-flying programmes structured, medically led and transparent, so that seeking help does not become a professional dead end.
Crew-pairing decisions belong in that system. Rosters are not merely a workforce-management issue, they shape fatigue, social isolation, sleep disruption and the likelihood that colleagues can recognise when someone is struggling. The 'Miami crew-pairing' findings should prompt airlines to review whether scheduling creates avoidable human-performance risk, rather than treating mental health as a problem located only inside an individual pilot.
Will IATA set a global standard? IATA is an industry association, not the global rule-maker. It can convene airlines, publish operational guidance, encourage best practice and use its audit and data-sharing tools to spread common standards. But mandatory international medical rules are set by national regulators and at the global level, through ICAO standards and recommended practices.
A meaningful outcome would be a common minimum framework rather than a universal psychological test. That framework could include an independently run peer-support programme, defined medical and safety escalation rules, fatigue-risk management connected to crew scheduling, protected reporting, specialist treatment access and a credible return-to-work pathway.
Such a model would recognise the central reality of aviation safety. Pilots are not the weak link in an otherwise perfect system. They are skilled professionals working inside systems that can either support good judgment or steadily erode it. In an age of automation, the most valuable safety technology may be the system that makes it easier for a human being to say, early and without fear, 'I need help'.