AI adoption UK is at a turning point. This week I facilitated the first AI Thinking Round Table, something I’d wanted to do for a while.

The idea was simple. An informal, online space for honest conversations about AI, with no slides, no presentations and no fixed agenda. Just people thinking out loud together about what AI actually means for their work and their organisations.

What came out of the conversation echoed so much of what I’m hearing from clients and networking contacts at the moment. The same questions, the same uncertainties, the same moments of recognition when someone names something that others have been quietly thinking but haven’t said out loud.

The group were professionals from IT staffing, finance, project management, and climate tech, and as it turned out, an all-female group. This reflected the AI adoption UK gender gap, which meant this reality wasn’t just a statistic we referenced. It was something people in the room had felt personally.

One moment I haven’t been able to stop thinking about. Someone mentioned feeling embarrassed to admit they use AI. Not because they thought it was wrong, but because of how it might be perceived.

That tells you something important about where we actually are with this.

AI adoption UK: what the room agreed on

AI isn’t going to slow down. The financial incentives are too strong and the pace too fast. The question isn’t whether, it’s how responsibly it happens.

The human side came through clearly too. Recruitment, customer service, professional services… these still need personal connection and discernment that AI can’t replicate. Not in the ways that matter most.

The governance point got the most reaction. Only 6% of corporate spend is going to AI governance, despite 88% of employees wanting to use AI tools. That gap, between what people are already using and what organisations are formally sanctioning, is where the real risk sits. Not in the technology. In the absence of guardrails around it.

If you want to understand how AI is changing the workplace, start with the people already using it.

What the room was less sure about

Jobs are transforming rather than disappearing, but the entry-level picture is worrying. Law firms and finance are already showing a 25% reduction in graduate hiring as research and analysis tasks get automated. Roles are evolving. The question nobody could fully answer was whether the people who would have had those roles get the chance to evolve with them.

There was also an honest conversation about brain atrophy, whether outsourcing thinking to AI gradually weakens our ability to think without it. The counterargument was that AI actually forces faster learning. I’m not sure either side convinced the other.

The thing I keep coming back to

Nobody last night was asking “should we use AI?” That conversation is over.

The questions were harder. Where does my judgement still matter? How do I filter AI output the way I’d filter advice from a trusted friend, taking it seriously but not treating it as the final word? What does AI adoption UK look like here, in my organisation, not in the abstract?

Those are the right questions. And they don’t get nearly enough airtime.

For more on how AI is reshaping how we do business, check out my Elaine Gold Consulting homepage.

Join the next conversation

The next Round Table date will be confirmed shortly. If you’d like to be in the room, the best way to hear about it is through my weekly newsletter, AI Made Useful. Practical, human, no hype.

Sign up here

And if any of this is sounding familiar, I’d love to hear from you. These are the conversations worth having.

The post The AI Thinking Round Table — what the first conversation revealed appeared first on Elaine Gold |AI + Human Connection.

“I’m using one AI platform for marketing, another for creating artefacts, another for research. I’m getting confused.”

AI tools for solopreneurs have multiplied fast — and so has the confusion around them.
I heard a version of this three times this week.

Different people, different businesses, different tools. But the same underlying problem: AI has become another source of complexity rather than a solution to it.

And the frustrating thing is that all three people were trying hard. They’d invested time and money. They weren’t resistant to AI — they were enthusiastic about it. They just had no clear system. And without a system, more AI tools means more confusion, not more productivity.

Which AI tools for solopreneurs actually work?

The AI industry — and much of the content around it — is tool-obsessed. Which platform is best. Which one does what. Which one you should switch to next.

That conversation is almost entirely unhelpful for someone trying to get real work done.

Because the question that actually matters isn’t which AI tool to use. It’s which task to start with.

What in your working week takes the most time? What’s repetitive, draining, and doesn’t need your full creative attention to do well? What could be done faster — or better — with some support?

Start there. One task. The one that causes the most friction or costs the most time. Then find — or identify from what you already have — the tool that fits that specific job.

That’s the starting point most people skip. And it’s the reason AI stays confusing rather than useful. The right AI tools for solopreneurs are not the newest ones — they’re the ones that fit a specific task.

Why using multiple AI tools is making things harder, not easier

There’s a particular pattern I see with solopreneurs and independent business owners who’ve been exploring AI for a while.

They’ve accumulated tools. One for writing. One for images or documents. One for research. Each one made sense at the time. But now they’re context-switching between platforms, maintaining different prompting styles, and spending more time managing the tools than doing the work.

The solution isn’t necessarily to stop using them all. It’s to be deliberate about which one earns its place in your regular workflow — and to build from there rather than adding more.

A focused AI workflow with one well-used tool will outperform a fragmented one with five every time. For AI tools for solopreneurs to genuinely save time, they need to fit into a routine — not sit on a list of things to explore.

AI productivity depends on accountability, not just the right tool

There’s something else worth naming this week, prompted by a piece in Harvard Business Review on what they’re calling “workslop” — AI-generated output that looks polished but contains no real thinking behind it.

It’s becoming a trust issue in teams and businesses. People start wondering whether anything they receive has actually been thought through.

The answer isn’t to stop using AI. It’s to stay accountable for what you produce. To know which outputs need a proper human review before they go anywhere — and to build that step into your workflow as a matter of habit, not an afterthought.

If it goes out under your name, it’s yours. The tool is not responsible for it. You are.

How to build a simpler AI workflow this week

If you’re feeling overwhelmed by AI tools right now, here’s a simple three-step reset:

Stop adding new tools. Work with what you have.

Pick one task — the one that takes the most time or causes the most friction. Apply one tool to that task only.

Build a review step in from the start. Not because the output will be wrong, but because the habit of checking protects your quality and your reputation.

That’s a more useful week than any amount of tool-testing.

Ready to build an AI workflow that actually fits your work?

If you’d like to work through this properly, book a free 15-minute discovery call

Or join the AI Thinking Round Table on 28 April, 7pm — free, informal, no agenda.

Register here

Sources: Harvard Business Review — Why People Create AI Workslop and How to Stop It

Elaine Gold | elaine-gold.com | AI + Human Connection

The post Too many AI tools and not enough clarity — how solopreneurs can find their starting point appeared first on Elaine Gold |AI + Human Connection.

It’s one of the most common conversations I have with SME leaders right now. The tools are in place. The licenses are paid for. And the team is still working exactly as they did before.

The instinct is to look at the technology — a different rollout, more training sessions, better comms. But in most cases, the tool isn’t the problem.

The problem is confidence. And confidence comes from clarity, not from more Copilot tutorials. Improving Microsoft Copilot adoption starts with addressing both.

Microsoft Copilot Adoption: What low adoption actually looks like

When I run Copilot workshops, I hear the same things at the start:

• “I’m not sure it can help with what I do.”
• “I tried it once and the output wasn’t great, so I stopped.”
• “I don’t really know where to start.”

These aren’t excuses. They’re honest responses from people who haven’t yet been helped to connect Copilot to the specific tasks that would make their working day easier.

That connection — between the tool and the work — is where adoption actually begins. And it rarely happens on its own.

The framework that shifts things

Whether I’m working with a team in a workshop or with an individual in a 1:1 session, the approach is the same. Three steps, in order.

1. Right tasks

Before anyone writes a prompt, the question is: which tasks actually belong to Copilot?

The ones that work best share a few characteristics. They’re repeatable. They’re time-consuming without needing deep judgement. And the output can be checked and refined before it goes anywhere.

A useful test: if Copilot got this slightly wrong, would you catch it before it caused a problem? If the answer is uncertain, that task needs more human input — not less.

Tasks that need your voice, your relationship context, or a call only you can make — those stay with you. Copilot can support the thinking. It shouldn’t replace it.

2. Right instructions

Most people who get poor output from Copilot aren’t doing it wrong. They’re being too vague.

Copilot needs context, not just a request. Format, tone, constraints, what you don’t want — the more precisely you brief it, the more useful the output. This is true however capable Copilot becomes at suggesting prompts or anticipating your needs. Precision in instructions still matters.

This is a learnable skill. It takes practice, not technical ability. Boosting Microsoft Copilot adoption in your team is largely about building this skill.

3. Review — always

Every piece of output that goes out under your name needs a human check. Not because Copilot is unreliable, but because you are accountable for what you produce.

A review step protects your quality, your voice, and your reputation. It doesn’t have to be long. It has to happen.

What changes in the room

By the end of a Copilot workshop, something shifts.

Not because participants have learned a long list of features. But because they’ve worked through a simple question: which of my tasks take the most time — and which of those don’t actually need my full attention to do well?

Once someone can answer that, Copilot stops being abstract. It becomes relevant to them, specifically. And that’s when they’re willing to try.

I’ve seen people go from “I don’t see how this helps me” to identifying three or four specific use cases they’re genuinely curious about — in 60 minutes. Not because the tool changed, but because their confidence did.

Two ways to move forward

For individuals: If you want to get consistent, useful results from Copilot — or from AI more broadly — a 1:1 working session gives you a clear, practical starting point. We work on your tasks, your instructions, and a review process that fits how you actually work.

For teams: If you lead a team and you’re watching licenses go unused, a workshop conversation is likely more useful. I work with teams to build confidence and practical capability — not through feature training, but through task clarity and hands-on practice.

Either way, it starts with a free 15-minute conversation.

Book a discovery call here

For more about me, check out my About page

The post Why your team isn’t using Microsoft Copilot — and what actually fixes it appeared first on Elaine Gold |AI + Human Connection.

If you’ve ever taken the Myers-Briggs Type Indicator, you’ll know your four-letter type. What’s less commonly discussed is how that type shapes the way you use — and misuse — AI.

I’m an MBTI practitioner with over 30 years of experience using personality frameworks in leadership development and organisational change. I’ve started this series because the question of who is sitting in front of the AI matters just as much as which tool they’re using.

This article looks at the NT temperament group: INTJ, INTP, ENTJ, and ENTP. If that’s your type, read on. The relationship between NT personality types AI and effective use is what this article explores.

What NT Personality Types AI Users Have in Common

The NT temperament is sometimes called the Rationals. These four types share a fundamental orientation towards logic, systems, and ideas. They’re drawn to complexity. They want to understand how things work, find the gaps in an argument, and build better models. These characteristics define how NT personality types AI interactions tend to unfold.

NTs tend to be:

• Strategic and conceptual thinkers
• Confident in their own reasoning
• Drawn to competence and intellectual rigour
• Comfortable with complexity and ambiguity
• Impatient with process for its own sake

These traits show up differently depending on whether you’re an introvert or extravert, and whether your preference is more for structured outcomes (J) or open exploration (P). An ENTJ moves fast and leads from the front. An INTP will explore an idea from every angle before committing. An ENTP debates their way to clarity. An INTJ builds the strategy and wants it executed properly.

But the underlying NT orientation — logic first, systems thinking, intellectual confidence — runs through all four.

Where NT types naturally do well with AI

Of all the temperament groups, NTs tend to adapt to AI fastest. And there are good reasons for that.

AI is genuinely well-suited to NT thinking. It can hold complexity, work through frameworks, generate and stress-test ideas at speed, and communicate nuanced thinking clearly. For someone who finds that kind of work energising, AI starts to feel less like a tool and more like a capable thinking partner.

NT types tend to use AI particularly well for:

Strategic thinking and frameworks. AI is good at helping you build out a model, test its logic, spot gaps, and stress-test assumptions. For NTs who think this way naturally, it accelerates something they’re already good at.

Research and synthesis. Pulling together information from multiple angles quickly, then making sense of it — AI handles this well, and NTs are well-placed to evaluate the output critically.

Writing and communication. Many NT types think in complexity but need to communicate simply. AI is useful for translating dense thinking into clear language — especially for audiences who don’t share your technical or strategic background.

Pressure-testing ideas. This is where AI is underused by most people but can be genuinely valuable for NTs. Ask it to argue against your position. Ask it where the logic breaks down. Used this way, it’s a real intellectual sparring partner.

The blind spot most NT types share

Here’s where I’ll be direct — and I say this as an INTJ myself.

The biggest risk for NT types using AI isn’t resistance or confusion. It’s overconfidence.

NTs trust their own thinking. That’s usually a strength. But AI is also confident. It produces fluent, well-structured output that looks right — and an NT mind will often read that output quickly, make a fast judgement, and move on.

Which means errors get missed. Not through carelessness, but because your brain is already three steps ahead.

The specific patterns I see most often:

Evaluating logic instead of checking facts. NT types are good at spotting whether an argument is structurally sound. But AI can be logically coherent and factually wrong. If you’re assessing the argument and not the underlying data, things slip through.

Confirmation through dialogue. NTs often use AI to develop and extend their thinking — which is valuable. But if most of your prompts are asking AI to build on what you already believe, you’re not getting the challenge you think you are. You’re getting a sophisticated echo.

Moving too fast through the review step. The instinct for most NT types is speed. Get to the answer, test it quickly, move on. The habit worth building is the deliberate pause — sitting with the output slightly longer than feels necessary before acting on it.

One practical shift that makes a real difference

I now routinely ask AI to challenge me rather than just help me.

Instead of “help me make the case for X,” I ask “what are the strongest arguments against X” or “where does this reasoning break down.” It changes the quality of what comes back — and it means I’m genuinely testing my thinking, not just accelerating it.

For NT types, that’s the most valuable shift. Not using AI as a faster version of your own brain, but as the thing that catches what your own brain is primed to skip.

This is where a 1:1 session goes deeper

Understanding your type is useful. Applying it to how you actually work — your decisions, your communication, your use of AI day to day — is where it gets practical.

In a 1:1 AI Clarity Session, we start with exactly this: how your preferences shape the way you’re currently using AI, where the gaps are, and what to change. It’s not generic AI advice. It’s specific to how you think.

If you’re an NT type who wants to use AI with more precision and less risk, that’s a good conversation to have.

Book a short exploratory call with me here

The post How NT Personality Types Can Get More From AI — And Where to Watch Out appeared first on Elaine Gold |AI + Human Connection.

Most conversations about adoption conversations focus on tools, platforms, and use cases. Very few talk about the person using them.

The truth is, your natural working style shapes how you approach AI — how quickly you adopt it, where you hesitate, what you do well with it, and where you might need support. Understanding that is more useful than any list of prompts.

This post is the first in a 16-part series exploring every MBTI personality type and its relationship with AI. It’s aimed at two audiences: individuals who want to use AI more confidently and effectively, and leaders and managers who want to support their teams through team adoption in a practical, human-centred way.

We start with the ISTJ.

  About this series

This series runs for 16 weeks, covering one personality type per week. The goal is not to put people in boxes — it’s to help individuals and organisations understand what good AI adoption actually looks like for different kinds of people. Non-technical, practical, and strengths-led throughout.

What Is MBTI and Why Does It Matter for AI?

The Myers-Briggs Type Indicator (MBTI) describes 16 personality types based on four dimensions: how people direct their energy (Introversion/Extraversion), how they take in information (Sensing/Intuition), how they make decisions (Thinking/Feeling), and how they approach structure (Judging/Perceiving).

These preferences shape how people respond to change, ambiguity, new tools, and new ways of working. AI adoption is all of those things at once. Which is why this dimension is one of the most underused lenses in any conversation about AI in the workplace.

I’m not suggesting personality determines what someone can do with AI. It shapes where they start, what feels natural, and where they might need a different kind of support. That’s a meaningful distinction — and a useful one for leaders managing teams through change.

The ISTJ Personality Type: A Brief Overview

ISTJ stands for Introverted, Sensing, Thinking, Judging. ISTJs are often described as reliable, methodical, and thorough. They work systematically, take their responsibilities seriously, and hold high standards for both themselves and their work.

In a professional context, ISTJs tend to be the people who deliver consistently, flag problems early, and do not cut corners. They are not always the loudest voice in the room, but they are frequently the most dependable one.

When it comes to AI, the ISTJ’s profile creates a specific and rather interesting pattern.

How ISTJs Tend to Approach AI Adoption

The initial hesitation

ISTJs do not typically rush to adopt new tools. They want to understand how something works, what it’s for, and what the rules are before they begin using it. This can look like resistance from the outside — but it isn’t. It’s due diligence.

In my experience working with leaders and their teams, ISTJs are often the last person to announce they’re using AI. They’re also frequently among the first to be using it well.

The natural advantages

Once an ISTJ engages with an AI tool, several of their core strengths come directly into play:

—  Structured thinking ISTJs write precise, specific prompts — because they think in briefs, not in vague requests. This is one of the most important skills in getting good results from any AI platform.

—  Quality instinct They review output carefully before it goes anywhere. In a world where AI errors are common and often missed, this is a significant professional advantage.

—  Habit formation When an ISTJ finds a prompt or workflow that works, they use it consistently and refine it over time. That kind of disciplined, repeatable practice is exactly how reliable AI use develops.

—  High standards They are not satisfied with output that is almost right. That dissatisfaction drives better results.

Where ISTJs sometimes get stuck

The main risk for ISTJs with AI is waiting too long to start. The absence of a clear policy, an approved use-case list, or explicit guidance can cause them to hold back longer than necessary. They are not being obstructive — they are being thorough. But they benefit from clear organisational signals that it is safe and appropriate to begin.

ISTJs tend to do best when tools sit inside systems they already use and trust. Introducing a new platform alongside a new way of working creates unnecessary friction. Starting within familiar software removes that barrier.

ISTJs tend to do best when AI sits inside tools they already use and trust. Introducing a new platform alongside a new way of working creates unnecessary friction. Starting within familiar software removes that barrier.

—  Microsoft Copilot (Word, Outlook, Excel) Ideal starting point. Copilot integrates directly into Microsoft 365, which many ISTJs already rely on. Use it to draft documents in Word, manage email in Outlook, and work with data in Excel without leaving your existing workflow.

—  ChatGPT Works well for ISTJs who write detailed, structured prompts. Provide clear context, format requirements, and examples for best results.

—  Claude (Anthropic) Strong for longer documents, policy drafts, and written content where tone and structure matter.

—  Google Gemini Useful for research tasks and summarising longer documents within Google Workspace.

—  Perplexity AI Good for fact-checking and research where sourced, verifiable answers matter — which they typically do for ISTJs.

  Practical tip for ISTJs

Write your AI prompt the way you’d write a proper brief. Include: what you need, the format, the audience, the length, what to include and what to leave out. That specificity is not extra work — it’s the skill.

What Leaders and Managers Should Know About ISTJs and AI

If you manage ISTJs, the most useful thing you can do is remove the ambiguity around AI use in your organisation. That means:

—  A clear, written AI usage policy that specifies what is and isn’t approved.

—  An approved list of use cases so they know where to start.

—  A process for raising concerns — because ISTJs will have them, and they’re usually worth hearing.

—  Explicit acknowledgement that reviewing and checking AI output is a valued part of the process, not an inefficiency.

ISTJs are not AI sceptics by nature. They are rigorous professionals. Give them the structure and the signal, and they will become consistent, high-quality AI users who help raise the standard of AI output across the team.

They are also the people who will catch the errors others miss. In any organisation where AI output is being used in communications, reports, or client-facing work, that capability is worth actively protecting.

The One Habit That Makes the Biggest Difference

For ISTJs specifically: start with one task, not AI in general.

Pick something specific — drafting a weekly report, summarising meeting notes, producing a first draft of a standard document — and build one reliable workflow around it. Get that working properly before adding anything else.

That approach suits the ISTJ’s natural working style and produces better results than broad exploration ever does. It also builds confidence incrementally, which is how ISTJs tend to work best.

Does personality type really affect how people use AI?

Yes — significantly. Personality shapes how people respond to ambiguity, new tools, and change. It affects whether someone jumps in or holds back, whether they experiment freely or prefer a structured starting point, and what kind of support helps them most. MBTI is one useful lens for understanding those differences at work.

What is the best AI tool for an ISTJ?

Microsoft Copilot is often the best starting point because it sits inside tools ISTJs already use — Word, Outlook, Excel — without requiring a new platform or workflow. ChatGPT and Claude also work well for ISTJs who write detailed, structured prompts.

How should managers support ISTJs who are slow to adopt AI?

Provide clear written guidance: a usage policy, an approved use-case list, and a process for raising questions. ISTJs are not resistant to technology — they are thorough. Remove the ambiguity and most of the hesitation goes with it.

Is MBTI a reliable framework?

MBTI is best used as a lens rather than a definitive assessment. It describes tendencies and preferences, not fixed categories. In the context of AI adoption, it helps explain why people approach new tools differently — and what kind of support is likely to help. It is not a prediction of what someone can or cannot do.

About This Series

MBTI × AI is a 16-week series by Elaine Gold, exploring how every personality type approaches AI adoption and use. Each post covers: how that type naturally approaches AI, where they excel, where they get stuck, practical platform tips, and what leaders can do to support them.

New posts publish weekly. The series covers all 16 MBTI types: ISTJ, ESTJ, INFJ, ENFJ, INTJ, ENTJ, ISFJ, ESFJ, ISTP, ESTP, ISFP, ESFP, INTP, ENTP, INFP, and ENFP.

Elaine Gold works with leaders, teams and individuals on practical, human-centred AI adoption. Her work focuses on AI + Human Connection — helping people use AI with confidence, judgement, and clarity.

The post MBTI and AI: How Your Personality Type Shapes the Way You Work with AI appeared first on Elaine Gold |AI + Human Connection.