Why CEOs Are Using ChatGPT as a Sounding Board, and What's Missing
5 May 2026
CEOs use ChatGPT for quick ideas, but it lacks company context, nuance and continuity; adviser-trained AI bridges the gap.
CEOs are turning to ChatGPT for quick, round-the-clock decision support. With its constant availability, many leaders use it to brainstorm ideas, draft documents, or even as a "therapist" for processing thoughts. However, while AI offers speed and convenience, it falls short in areas like company-specific context, nuanced judgement, and handling novel challenges.
Key Points:
AI’s Role: 64% of executives use or plan to use generative AI for decision-making.
Trust Issues: 74% trust AI for advice over colleagues, but 93% admit AI has led to errors.
Limitations: AI lacks memory of company history, struggles with creative thinking, and often provides generic solutions.
Human vs AI: While AI excels at data processing, human advisers bring deeper understanding and consistency.
Solution:
Custom AI models, trained with specific business data and adviser expertise, bridge these gaps. They combine AI’s speed with the tailored insights of human advisers, ensuring better decision-making and leadership support.
Takeaway: ChatGPT is a helpful tool, but it works best when paired with human expertise to ensure decisions are both fast and well-informed.
Where ChatGPT Falls Short on Leadership Judgement
Generic Answers vs Company-Specific Context
ChatGPT starts fresh with every session, meaning it doesn’t retain a structured understanding of your business. Your industry, pricing model, competitors, and strategic goals all need to be reintroduced every time. This can feel like you’re repeatedly explaining your company from scratch.
"Memory is not context. Knowing that 'you work at a SaaS company' is not the same as having a structured business profile with your industry, target market, competitors, pricing model, differentiators, and strategic objectives all organised." – Fluxel
Without this persistent context, leaders are left to manually reconcile new AI insights with previous advice. Under pressure, it’s easy to miss inconsistencies across multiple AI-generated recommendations. This gap in continuity directly impacts the depth of leadership judgement the tool can provide.
Missing Nuanced Judgement
While ChatGPT can deliver well-researched answers, it often lacks the creative and critical thinking that leadership requires. Dean McKeown, Director of AI at Smith School of Business, captures this limitation:
"Important variables in good leadership - like creativity, thinking outside the box and critical thinking - may not be apparent in what a chatbot has to say."
The AI struggles most with uncharted challenges. Since it depends on patterns in existing data, it’s unable to craft original solutions for unprecedented situations, such as unexpected tariff changes, supply chain crises, or pivotal brand moments. Research involving 640 entrepreneurs revealed a striking contrast: while AI improved profits by 10% to 15% for high-performing businesses, it caused an 8% decline for struggling entrepreneurs - largely because they lacked the judgement to filter out generic or unsuitable advice.
A side-by-side comparison helps illustrate where ChatGPT falls short compared to human advisors.
Comparison Table: AI vs Human Advisor Judgement
Feature | ChatGPT | Human Advisor |
|---|---|---|
Context awareness | Processes inputs without understanding company history or culture | Integrates a nuanced understanding of team dynamics and stakeholder priorities |
Consistency | Prone to "forgetting" key assumptions in long-form sessions | Maintains logical continuity and a long-term strategic vision |
Error handling | Offers "my bad" without explaining the logic of the mistake | Provides reasoning for errors and builds trust through transparent analysis |
Decision support | Best for brainstorming and short-turn tasks | Essential for high-stakes judgement and for distinguishing good ideas from mediocre ones |
These limitations underline the risks of relying solely on AI for strategic leadership. While ChatGPT can be a useful tool, its shortcomings highlight why human expertise remains irreplaceable in making informed, high-stakes decisions.
The Costs of Depending Only on Generic AI
Relying too heavily on generic AI can undermine the depth and continuity of trusted advisory relationships, which are essential for effective leadership.
Weakening High-Trust Advisory Relationships
Strong CEO-adviser relationships thrive on continuity and a deep understanding of context. But leaning on generic AI for quick answers can chip away at this foundation. AI lacks the ability to grasp the nuances of your organisation or the relational context that builds trust in advisory work.
A 2025 study highlights a concerning trend: many executives are turning to AI for decision-making, reducing their engagement with long-standing human advisers. This shift means advisers, who typically have a comprehensive understanding of strategic goals, team dynamics, and stakeholder needs, are relegated to formal sessions rather than being part of the leader's day-to-day decision-making process.
"The work that matters most right now - the work that AI cannot do - is exactly the work executive coaching has always been about... building the trust that makes people follow you through change." – Michael Rolph, Executive Coach
This erosion of trust poses a significant risk. Leaders may place undue confidence in AI outputs, potentially compromising the quality of their decisions.
The Danger of Overconfidence in AI Responses
Over-reliance on generic AI can lead leaders to trust outputs that lack the critical human judgement needed for sound decisions. The confidence gap is stark: while 70% of C-suite executives describe themselves as "very confident" in AI, only 27% of intermediate staff trust their leaders’ ability to use AI effectively. This disconnect goes beyond perception - 93% of top executives admit to making decisions based on AI outputs that stemmed from inaccurate data, and 40% report that these errors have already caused significant business problems. Alarmingly, 78% of executives use AI for tasks they aren’t trained to manage, leading to misplaced trust in potentially flawed results.
In October 2025, a study by Immerss, a consultative commerce platform, compared pure AI chatbots with a hybrid model for high-value jewellery purchases (£4,000–£8,000). The pure AI approach achieved only a 3% conversion rate and a 28% return rate. When they switched to a hybrid model - where AI handled initial filtering and humans managed final consultations - conversion rates soared to 24–31%, and customer satisfaction reached an impressive 98%.
"AI gives confidently wrong answers... High-value customers get generic, useless responses." – Rita Kalashnik, Consultative Commerce Expert
The takeaway is clear: overestimating AI’s capabilities can lead to initial enthusiasm but eventually results in dissatisfaction and costly corrections when its limitations become evident.
Adviser-Trained AI for Leadership Continuity
Advisers are now turning to tailored AI models to extend their expertise beyond scheduled sessions, offering continuous support that bridges the gaps where generic AI often falls short.
Making Adviser Expertise Available Between Sessions
In September 2025, Executive Coach Richard Hughes‐Jones teamed up with Rider Latham, CEO of Solo & Secret Spa, to create "Rider's AI Coach." This custom AI was built using a detailed questionnaire covering Rider's lifestyle and exit ambitions, coaching transcripts, and 360-degree feedback. The result? A system that both the coach and CEO could use to spot patterns and stay accountable to long-term goals between their regular sessions.
Similarly, in early 2026, Max Razmakhin, founder of Resident, launched a 24/7 AI business coach using a three-phase process. It started with 60 targeted questions, followed by uploading critical business documents like marketing plans, organisational charts, and investor decks. This ensured the AI's recommendations were deeply rooted in his company’s specific context, avoiding generic advice.
These models are trained across three key layers: business context, leadership profile, and a "coaching avatar" that mirrors the adviser’s style. Coaches can even upload their published works and proprietary frameworks, embedding their unique methods into the system. Crucially, these AI systems include guardrails - rules that compel the AI to challenge the user, disagree when needed, and highlight blind spots instead of simply providing agreeable responses.
This personalised, always-available support blends the speed of AI with the depth of human expertise, offering a seamless extension of the adviser’s role.
Combining AI Speed with Human Insight
When you combine the rapid processing power of AI with the nuanced, context-rich insights of human judgement, you get a powerful partner for decision-making. These AI systems are not just fast - they are tailored to align with your adviser’s specific guidance and your organisation’s strategic priorities.
"The biggest unlock in my personal growth this year wasn't a book, a retreat, or a coach. It was a GPT thread."
– Rider Latham, CEO, Solo & Secret Spa
Michael Hyatt, Founder of Full Focus, shares a similar sentiment:
"I use it between my coaching sessions to refine ideas, challenge assumptions, and set the stage for deeper coaching sessions. It's like having a sparring partner before the real match."
Starting the Conversation with Your Adviser
If you’re intrigued by the idea of personalised AI support, start a conversation with your adviser. Discuss how integrating a tailored AI model could provide continuous, context-aware assistance while preserving the trust and depth of your existing relationship. It might just be the next step in enhancing your decision-making and leadership journey.
Wrapping Things Up
In this article, we’ve explored how, while AI can deliver quick responses, it often falls short of offering the kind of deep, strategic insight that only a human adviser can provide. Sure, AI is fast, but it doesn’t grasp the intricacies of your business, your board’s dynamics, or the frameworks your adviser uses. And when decisions carry weight, that understanding is critical.
This gap highlights the importance of maintaining continuity in your advisory relationships. The question isn’t whether to use AI - it’s about ensuring that any AI tool you use complements the nuanced judgement of your trusted adviser, rather than trading depth for speed. Tools like GuidanceAI address this by integrating your adviser’s unique approach, ensuring you’re supported even between sessions.
As GuidanceAI aptly puts it:
"Doing nothing doesn't preserve the relationship. It quietly reduces the role your guidance plays in day-to-day decisions."
– GuidanceAI
If you’re already working with a coach or senior adviser, it’s worth having a conversation about how they’re approaching continuity. By doing so, you can ensure your decisions are backed by both speed and the thoughtful insight your adviser brings to the table.
FAQs
When is ChatGPT safe to use for CEO decisions?
ChatGPT can be a helpful tool for CEOs when used to gain quick insights, brainstorm ideas, or summarise information. However, it should never substitute human judgment, especially for complex or sensitive decisions. To minimise risks, it's essential to maintain robust governance and implement strong security measures.
How can I stop AI advice being too generic for my business?
To get more useful advice from AI, craft prompts that are detailed and packed with context. Tailor them to reflect your business's specific challenges, industry quirks, and goals. Vague instructions often lead to generic, less helpful outputs.
Combine AI's insights with human expertise to fill in gaps like nuanced decision-making and personalised context. This approach ensures the advice you receive is practical and aligns closely with what your organisation actually needs.
What does a coach-trained AI need to work well?
A coach-trained AI must meet six crucial standards to function effectively and responsibly:
Effective Communication: The AI should convey information clearly and understand user input to foster meaningful interactions.
Co-Creating Relationships: It should build trust and collaboration, enabling users to feel supported and engaged.
Facilitating Learning and Growth: The AI must encourage users to develop skills, insights, or strategies for personal or professional growth.
Promoting Ethical Practices: Upholding integrity and ethical guidelines is essential to ensure interactions align with moral principles.
Ensuring Privacy: Safeguarding user data is non-negotiable, maintaining confidentiality throughout all engagements.
Maintaining Accessibility: The AI should be designed to accommodate diverse users, ensuring inclusivity and ease of use.
These standards work together to enable the AI to support informed, responsible decision-making while respecting user needs and values.