Imagine an airline is preparing for a busy day.

Hundreds of flights are scheduled.

Thousands of passengers are travelling.

Weather conditions are changing.

Aircraft are moving between airports.

Crew schedules are being coordinated.

Behind all of this, artificial intelligence is quietly analysing information and identifying patterns.

An AI system produces a recommendation.

It looks convincing.

The data looks convincing.

The system has processed far more information than a human could reasonably analyse in the same amount of time.

So someone asks a very simple question:

Can we trust it?

That question may become one of the most important questions aviation has to answer as Artificial Intelligence becomes more deeply integrated into airline operations.

The issue is not whether AI can be useful.

It clearly can.

The bigger question is:

How do airlines make sure AI is being used safely, responsibly and with the right human oversight?

That is where AI Governance in aviation comes in.

AI Is Already Becoming Part of Aviation’s Future

Artificial Intelligence is no longer just something we hear about in technology conferences. It is becoming part of the aviation industry.

Airlines and aviation organisations are exploring how AI can be used for areas such as flight operations, safety, air traffic management, aircraft maintenance and other parts of aviation.

The International Civil Aviation Organization (ICAO) recognises both the benefits and risks of using AI in aviation. It says AI needs to be developed and used in a way that is safe, transparent, secure and accountable, with humans still playing an important role.

In Europe, the European Union Aviation Safety Agency (EASA) is also working on how AI can be used safely in aviation. In June 2026, EASA released an updated AI Concept Paper that looks at newer AI technologies and more advanced forms of automation. Its approach focuses on making AI trustworthy while keeping aviation safety and human involvement at the centre.

So this conversation is not simply about whether airlines might use AI one day.

It is about how aviation can responsibly integrate increasingly capable AI systems.

And that brings us to five questions.

1. Is the AI Using Accurate Data?

This might sound like a technical question.

It is actually one of the most important governance questions.

AI is only as reliable as the information it is working with.

Imagine an AI system is helping an airline analyse operational information.

The system receives information from several different sources.

But what if some of that information is incomplete?

What if records are outdated?

What if information has been entered incorrectly?

What if two systems contain conflicting information?

The AI may still produce an answer.

And the answer may even sound extremely confident.

But confidence does not equal accuracy.

This is why Data Governance becomes such an important part of AI Governance.

Organisations need to understand where data comes from, whether it is fit for purpose, how it is managed and who is responsible for it.

ICAO has specifically highlighted the importance of information and data management for AI applications, noting that datasets used to train and validate AI need to be complete, representative and of high quality.

This is a lesson I find particularly interesting as I learn more about AI Governance.

Before asking:

What can the AI do?

we should also ask:

What information is the AI learning from?

2. Can a Human Challenge the AI?

Now imagine the AI has analysed the data and produced a recommendation.

Should the airline simply follow it?

Not necessarily.

Aviation has always relied heavily on human expertise, procedures and professional judgement.

AI does not change that overnight.

Instead, the challenge is finding the right relationship between humans and AI.

EASA’s work on Level 2 AI specifically addresses human AI teaming, where AI systems can perform certain decision making functions while remaining under human oversight. EASA also emphasises the importance of safe human AI interaction.

In simple terms:

The AI can make a recommendation, but the human needs to understand it, evaluate it and have the ability to intervene when appropriate.

Think of the AI as an extremely sophisticated alarm clock.

It can wake you up.

It can tell you something needs attention.

But it should not automatically decide what you do next.

That distinction matters enormously when AI is used in aviation.

3. Can We Understand Why the AI Made Its Recommendation?

Imagine an AI system tells an airline:

There is a high risk associated with this situation.

A natural question follows:

Why?

If nobody can explain what information influenced the recommendation, it becomes difficult to properly assess the result.

This does not mean every AI system needs to reveal every mathematical calculation behind its output to every user.

It means organisations need appropriate levels of explainability and transparency for the context in which the AI is being used.

This becomes particularly important when AI is involved in safety related applications.

EASA’s aviation AI work includes concepts such as AI explainability and learning assurance as part of developing trustworthy AI systems.

The question therefore becomes:

Can the people responsible for the decision understand enough about the AI’s output to use it safely?

If the answer is no, that itself becomes a governance concern.

4. Who Is Accountable When AI Gets It Wrong?

This may be one of the most uncomfortable questions.

Imagine an AI system makes a recommendation.

A human follows the recommendation.

Later, the decision turns out to have been wrong.

Who is responsible?

The AI?

The person who approved the recommendation?

The airline?

The team that developed the system?

The organisation that supplied the technology?

This is why AI Governance cannot simply be about whether the technology works.

It must also establish accountability.

People need to understand their responsibilities.

There should be appropriate processes for monitoring the system.

There should be mechanisms for reporting problems.

And there should be clarity around who has authority to intervene.

ICAO has identified accountability, fairness, human control and technical robustness among principles relevant to trustworthy AI in aviation.

This is important because saying:

The AI made the decision cannot become a substitute for human accountability.

5. How Do We Know the AI Is Still Working as Intended?

There is another trap that is easy to overlook.

An AI system can be tested before deployment.

But that does not necessarily mean the work ends there.

The environment around the system can change.

Data can change.

Operational conditions can change.

The system itself may be updated.

New risks can emerge.

A model that performed well in one environment may behave differently when circumstances change.

That means responsible AI Governance needs to consider ongoing monitoring.

Is the system still performing as expected?

Are there unusual outputs?

Are errors increasing?

Are people becoming too dependent on the AI?

Is the system being used for something it was never designed to do?

These are governance questions that continue long after an AI system has been deployed.

ICAO has also highlighted continuous monitoring, data quality, explainability and human involvement as important elements of trustworthy aviation AI.

What Happens If the AI Is Wrong?

Let’s return to our airline.

The AI has made a recommendation.

The recommendation looks reasonable.

But a human notices something unusual.

Instead of automatically accepting the recommendation, the person investigates.

They check the underlying information.

They consider the operational context.

They question the AI output.

They may decide to accept it.

Or they may override it.

That is what meaningful human oversight can look like.

The goal isn’t to make humans ignore AI.

And it isn’t to make humans blindly follow AI.

It is to create a relationship where AI supports human decision making without removing appropriate human responsibility.

AI Governance Is Not About Stopping Innovation

Sometimes governance gets a bad reputation.

People hear the word “governance” and imagine paperwork, restrictions and endless approvals.

But good governance should not exist simply to prevent organisations from using new technology.

It should help organisations understand how to use that technology responsibly.

Think about an airline introducing an AI system.

Without governance, the question might be:

Can we deploy this?

With governance, the questions become more sophisticated:

What is this AI supposed to do?

What could go wrong?

What data does it need?

Is that data reliable?

What happens when the system is wrong?

Who monitors it?

Who can override it?

Who is accountable?

How do we protect the information it processes?

Those questions do not necessarily stop innovation.

They can make innovation safer.

Why Data Governance and AI Governance Cannot Be Separated

This is something that has become increasingly clear to me as I learn more about these areas.

You cannot have responsible AI without thinking about data.

AI needs data.

If the data is poor, the AI can produce poor results.

If the data contains personal information, privacy becomes important.

If access to the data is poorly controlled, cybersecurity becomes important.

If nobody knows who is responsible for the data, governance becomes a problem.

This creates a chain:

Data → AI → Risk → Privacy → Security → Governance

And aviation sits right in the middle of all of it.

That is why the future of aviation in the age of AI will require people who understand more than just the technology.

It will require people who understand risk, data, people and accountability.

What About Nigeria and Other Aviation Markets?

This conversation is not limited to Europe or the United States.

Airlines and aviation organisations around the world are moving into an increasingly digital environment.

That includes aviation markets across Africa, including Nigeria.

For countries developing their aviation technology capabilities, AI can create opportunities to improve efficiency, analyse information and support aviation operations.

But the governance questions remain relevant regardless of geography.

Is the data accurate?

Is passenger information protected?

Can humans challenge the AI?

Is the system transparent enough for its intended use?

Who is accountable?

AI Governance therefore needs to be considered in a way that recognises both international aviation principles and the regulatory and operational realities of individual countries.

The future of aviation will be global.

The governance of its AI will need to be thoughtful enough to operate within that global environment.

What Aviation Professionals Should Understand About AI

You do not have to be a software engineer to understand why AI Governance matters.

Aviation professionals already work with concepts that translate surprisingly well into responsible AI.

Risk assessment.

Procedures.

Human factors.

Incident reporting.

Safety culture.

Escalation.

Accountability.

Continuous improvement.

These ideas existed in aviation long before ChatGPT and Generative AI became household terms.

AI is introducing a new layer to the conversation.

The challenge is learning how to apply those established principles to increasingly intelligent systems.

EASA’s research into AI ethics found that aviation professionals have expressed concerns around AI performance, data protection and privacy, accountability and possible safety implications. The research also highlighted concerns about overreliance on AI and potential loss of human skills.

That tells us something important.

The conversation about aviation AI is not simply:

Can AI do this?

It is also:

Should AI do this, and under what conditions?

The Future May Not Be Human Versus AI

I don’t think the most interesting question is whether AI will replace humans in aviation.

The more interesting question is:

How will humans and AI work together?

AI can process enormous amounts of information.

Humans bring experience.

AI can identify patterns.

Humans can interpret context.

AI can raise alerts.

Humans can investigate them.

AI can make recommendations.

Humans can challenge them.

This is not about making AI powerless.

It is about making sure that increasing AI capability comes with appropriate oversight, accountability and trust.

What I Am Learning About AI Governance

As I continue my own transition into technology and learn more about AI Governance, Data Governance, cybersecurity and AI Privacy, I am beginning to see governance differently.

It is not simply about creating policies.

It is about asking difficult questions before problems occur.

And aviation is a perfect example.

When the consequences of an incorrect decision can be significant, we cannot simply assume that because an AI system is sophisticated, it is automatically trustworthy.

We need evidence.

We need controls.

We need monitoring.

We need good data.

We need human oversight.

And most importantly, we need clear accountability.

Frequently Asked Questions

Can airlines use AI for safety decisions?

AI can support aviation safety and operational decision making by analysing data, identifying patterns and providing recommendations. However, the appropriate level of automation and human oversight depends on the specific application, its risk and applicable aviation requirements.

What is AI Governance in aviation?

AI Governance in aviation refers to the policies, processes, controls and accountability mechanisms used to ensure AI systems are developed and deployed responsibly, safely and appropriately.

Why is human oversight important in aviation AI?

AI systems can produce incorrect or unexpected outputs. Human oversight provides a mechanism for qualified people to evaluate AI recommendations, intervene when necessary and maintain accountability.

Why does data quality matter for aviation AI?

AI systems depend on data. Inaccurate, incomplete or unrepresentative data can affect AI performance. This makes data quality and Data Governance important parts of responsible AI implementation.

Is AI going to replace aviation professionals?

Not necessarily. AI may automate or support specific tasks, but aviation still requires human expertise, judgement, accountability and oversight. The future is more likely to involve different forms of human AI collaboration across aviation.

On A Final Note

The question is not whether airlines should use Artificial Intelligence.

AI is already becoming part of the future of aviation.

The bigger question is how we use it.

An AI system can analyse thousands of pieces of information.

It can identify patterns.

It can raise an alarm.

It can recommend an action.

But when the decision matters, we still need someone willing to ask:

Are we sure?

That question is not a rejection of technology.

It is a sign of responsible technology use.

The future of aviation will probably involve increasingly capable AI systems.

But trust will not come simply because an algorithm is sophisticated.

Trust will come from good data, appropriate controls, transparency, human oversight, continuous monitoring and clear accountability.

And perhaps that is the real role of AI Governance in aviation.

Not to stop the aircraft from moving into the future.

But to make sure we know who is responsible for steering when AI comes on board.

Continue Reading

What Happens When AI Gets an Aviation Decision Wrong? Understanding Human Oversight in Aviation AI, What Happens When AI Gets an Aviation Decision Wrong? Understanding Human Oversight in Aviation AI

Explore why humans still need to review, challenge and make important decisions when AI is introduced into aviation.

Can AI Make Flying Safer? Here’s the Role of AI Governance in Aviation, Can AI Make Flying Safer? Here is the Role of AI Governance in Aviation

Discover how AI Governance can help airlines explore the benefits of artificial intelligence while keeping safety, accountability and human oversight at the centre.