Before You Automate Anything, Teach AI How You Communicate

If you spend enough time around aviation right now, you’ll hear a lot of extremes about AI.

Some people think it’s overhyped. Others think it’s going to replace entire industries overnight.

The reality is somewhere in the middle.

In business aviation and FBO operations specifically, AI is probably not going to replace the human side of the business. If anything, the opportunity is the opposite.

The best use of AI in aviation will be helping people spend less time fighting dated systems and processes, giving them more time serving customers.

Because at the end of the day, this is still a customer service/hospitality business.

AI Is Moving Fast

The challenge is that AI is changing quickly, the models are evolving at a very fast pace.

A bad experience from a year ago, or even 3 months ago may not reflect what these tools can do today. In some cases, capabilities that feel normal now either did not exist, were much weaker, or required more technical effort just a few months ago.

We have heard from a lot of people that don't use AI daily that formed their opinion of AI after one early interaction.

They asked a vague question. Got a generic answer. Decided it was overhyped and moved on.

While that may have been a fair reaction at the time (probably not), it is definitely not a fair read today.

The tools are improving quickly, and the way we use them has to evolve just as quickly.

The Industry Is Already Using AI

During recent industry discussions and events, one thing has become clear:

More people are using AI but there is a LOT of room for improvement. Even today most users lean on short prompts, minimal context and then they get generic outputs. That’s usually where frustration starts. Even heavy users are not harnessing the full capability available.

What’s interesting is that most people are still using AI like a search engine. And those that aren't, most of what we've found for use cases are:

• writing emails

• summarizing documents

Even at a more basic level, there are move advanced use cases you should try:

• research

• operational analysis

• marketing content

• spreadsheets

• financial analysis

• policy/SOP drafting

• customer communication

The people getting the best results are approaching AI differently.

They are treating it more like:

• a junior analyst

• a writing assistant

• a research partner

• an operational collaborator

And the difference in output quality is massive.

The Real Opportunity for FBOs

First thing people worry about is jobs, the opportunity is not removing people. In a service industry that is perpetually short staffed and highly physical, those worries are simply overblown.

The opportunity is removing repetitive administrative friction that frees up staff and resources to dedicate to the customer.

Things like:

• repetitive customer emails and calls

• managing SOPs, your policies make a great chatbot.

• shift summaries

• financial analysis

• operational reporting

• manual reconciliation (did someone say nightly closing!?)

• policy drafting

• internal communication cleanup

• repetitive documentation work

None of those activities are what make customers love an FBO.

Customer service does, hospitality does, the experience does!

Operational execution and creating customer experience is what drives revenue!

If AI can reduce repetitive workload by even 10–20%, that creates more time for:

• customer interaction

• ramp awareness

• training

• operational focus

• leadership development

How important is that in an industry where staffing pressure continues to grow?

One of the Most Useful AI Exercises We’ve Done

One of the more interesting internal exercises we performed at X-1 involved teaching AI how we actually communicate.

Not just what we say. How we say it.

We had AI analyze users:

• emails

• operational communication

• leadership messaging

• writing cadence

• vocabulary usage

• sentence structure

• paragraph pacing

• formatting habits

• communication patterns

The goal was not imitation.

The goal was context.

Once AI understood:

• tone

• structure

• vocabulary

• operational style

• communication philosophy

…the output quality improved dramatically.

Not because the AI became “human.”

Because the instructions became specific. This is one of the key lessons for new AI users.

Better Context = Better AI

Most weak AI output comes from weak input.

For example:

Weak Prompt

Write a customer email explaining our weather policy.

That usually produces generic corporate language.

Now compare that to:

Better Prompt

Write a short response to an aircraft owner following a service failure during a weather event (lightning) at our FBO.

Tone should be calm, direct, operationally transparent, and hospitality-focused. Acknowledge the issue clearly, avoid corporate filler or excuses, and focus on accountability and restoring confidence. Ask me 5 questions about the incident and our policies to help create the email.

Note the ask me 5 questions line. This is incredibly important when you start using AI as you build the context you need. Make a habit out of having AI ask you questions about the task you assigned it. It also important you ask it to challenge your decisions like a colleague would.

When you do this you get completely different resulta.

The more context AI has:

• the better it performs

• the more useful it becomes

• the more natural the output feels

  
  

How to Analyze Your Own Communication Style

Here is a simplified version of the type of prompt you can use with modern AI platforms like ChatGPT, Claude, Gemini, or Copilot.

Gather Your Communication Samples

You have a few ways to do this.

The easiest option is to use Microsoft Copilot if your company uses Outlook. Ask Copilot to analyze your sent emails and identify your writing patterns.

You can also use the AI tool you already use most often, such as ChatGPT, Claude, Gemini, or Copilot, and paste in examples of your own writing.

Good samples include:

• sent emails

• customer responses

• internal messages

• proposals

• meeting follow-ups

• operational updates

  
  

Try to use 10–20 real examples if possible. The more representative the samples are, the better the profile will be.

Copy and paste the highlighted text as your prompt.

Writing Analysis Prompt

Analyze the following writing samples, sent emails, or communication examples for deep linguistic, lexical, structural, and statistical communication patterns.

I want analysis of:

TONE & COMMUNICATION STYLE

• conversational tone

• executive tone

• operational realism

• emotional neutrality vs expressiveness

• confidence level

• directness vs hedging

• persuasion style

• collaboration style

• leadership framing

• conflict handling style

STRUCTURE & PACING

• sentence length distribution

• sentence fragment frequency

• paragraph length distribution

• pacing and cadence

• greeting and closing behavior

• use of bullets

• use of rhetorical questions

• transition habits

• formatting tendencies

• scanability patterns

VOCABULARY & LEXICAL ANALYSIS

• vocabulary complexity

• vocabulary depth

• lexical density

• abstraction level

• technical terminology frequency

• operational terminology usage

• industry shorthand usage

• conversational compression patterns

• preferred verbs

• preferred adjectives

• preferred filler words

• repeated phrases and recurring language structures

• tendency toward concrete vs abstract wording

• use of financial, operational, or technical framing

• frequency of jargon vs plain language

• phrase economy and information compression

PATTERN ANALYSIS

• average sentence length

• median sentence length

• percentage of fragments

• passive vs active voice usage

• readability scoring

• common sentence starters

• frequency of questions

• punctuation patterns

• average paragraph size

• word repetition frequency

• vocabulary uniqueness

Then generate:

1. Executive summary of communication style

2. Detailed writing behavior profile

3. Vocabulary and lexical profile

4. Statistical communication profile

5. AI instruction set for replicating the style

6. Common failure modes when attempting to imitate this style

7. A persistent context prompt for future AI sessions

Focus on observable linguistic behavior rather than personality assumptions.

The goal is not literary imitation.

The goal is operationally authentic communication that feels human, executive-level, and contextually believable.

What To Do With The Results

Having the results isn't going to help much unless you use them.

The analysis itself is not the value, the value comes from reusing the results.

ChatGPT

Best For

• ongoing conversational use

• email drafting

• iterative refinement

• maintaining long-term style consistency

How To Save Your Writing Style

Custom Instructions

In ChatGPT:

1. Click your profile icon

2. Go to Settings

3. Select Personalization

4. Open Custom Instructions

Paste:

• your communication profile

• tone guidance

• formatting rules

• vocabulary preferences

• what to avoid

This gives ChatGPT persistent context across future conversations. If your output is too long, you can always ask AI to summarize it.

Claude

Best for:

• long-form reasoning

• strategic documents

• financial analysis

• operational writing

• large context workflows

Recommended Approach

Use:

• Projects

• Project Knowledge

• Styles

Create a dedicated project for:

• management writing

• proposals

• operational communication

• customer communication

Upload:

• writing examples

• style guides

• communication profiles

• reference documents

Claude performs extremely well when persistent context, examples, and structured instructions are combined inside Projects.

Here is a prompt to help Claude create this from your communications analysis: Claude Communications Prompt

At the start of important sessions, paste:

• the full profile

  or

• the condensed version

Claude handles large contextual instructions extremely well.

Claude Tip

Claude performs extremely well when this information is combined inside Projects:

• real examples

• persistent context

• contrast instructions

• “avoid this / prefer this” guidance

The more examples you provide, the more natural the outputs become.

Microsoft Copilot

Best For

• Outlook

• Teams

• Word

• Excel

• enterprise workflow integration

How To Save Your Writing Style

Copilot is more task-oriented and less conversationally persistent.

Best approach: Create a reusable “AI Writing Profile” document and store it in:

• OneDrive

• SharePoint

• Teams files

• Word

Then paste portions of it into prompts when drafting:

• emails

• meeting summaries

• reports

• customer communication

Important Copilot Tip

Copilot strongly defaults toward:

• HR language

• legal-safe language

• corporate phrasing

You usually need to explicitly instruct it to:

• be concise

• sound conversational

• avoid corporate filler

• prioritize operational clarity

Otherwise outputs tend to sound generic.

Gemini

Best for:

• Google Workspace

• Gmail

• Docs

• meeting summaries

Recommended Approach : Maintain a reusable prompt document inside Google Docs and reuse it frequently.

Gemini benefits from repeated tone reinforcement and explicit formatting instructions.

Best For

• Gmail

• Google Docs

• Workspace environments

• meeting summarization

How To Save Your Writing Style

Gemini currently benefits from repeated reinforcement.

Recommended approach: Create a reusable “Communication Style Prompt” inside Google Docs.

Store:

• writing profile

• tone instructions

• examples

• formatting rules

Then paste relevant sections into new sessions when needed.

Gemini Tip

Gemini responds better when instructions are:

• explicit

• concise

• repeated consistently

Without reinforcement, style consistency can drift over time.

AI Still Requires Human Judgment

Personalization creates some risks, being aware of them can help mitigate them.

One of the risks with highly personalized AI output is that it can sound extremely convincing.

Especially when:

• tone matches

• vocabulary matches

• formatting matches

• decision framing matches

That does NOT guarantee:

• accuracy

• operational correctness

• compliance

• factual validity

AI-generated content should still be reviewed like work produced by a junior employee:

• useful

• fast

• often insightful

• but not automatically authoritative

The better AI becomes at sounding human, the more important human verification becomes!

Universal Best Practice

Regardless of platform, maintain three reusable assets:

1. Communication Style Profile

Short summary of:

• tone

• vocabulary

• formatting

• pacing

• communication philosophy

Usually 1–2 pages maximum.

2. Writing Example Library

Real examples of:

• emails

• operational communication

• proposals

• customer responses

• leadership messaging

AI learns extremely well from examples.

3. Daily Reusable Prompt

Simple instructions reused frequently.

Use the communication profile already established.

Prioritize:

• concise structure

• operational realism

• conversational executive tone

• direct communication

• short paragraphs

Avoid:

• corporate filler

• excessive hedging

• marketing language

• generic AI phrasing

The Most Important Takeaway

AI does not automatically understand:

• your business

• your tone

• your customers

• your communication style

• your operational philosophy

You have to teach it.

The organizations getting the most value from AI today are usually not the most technical.

They are the ones giving AI:

• context

• examples

• constraints

• feedback

• operational perspective

Real leverage starts once you've help AI understand you and your goals.

And in business aviation, that leverage should ultimately create something more human, not less: better service, better communication, and better operational awareness.