AI Thinking for the CEO Pragmatist

I recently flicked to a VentureBeat article by "thought leader" Khufere Qhamata which I thought would shed some insight on "AI Thinking", which I regard as the true intellectual property of training your organisation in "AI". That is, not how to use AI tools, strategise or write an AI Ethics Policy, but how to think generative "AI First" in how you approach your daily work in every role. As Peter Brand said in Moneyball, "this is getting things down to one number".

D-i-s-a-p-p-o-i-n-t-e-d.

Apparently, "The future belongs to those who understand how to think and communicate in vectors...with understanding how to translate human insight into the language of vectors and patterns that AI systems understand".

Seriously? It's no wonder that CEOs tune out on the possibilities of AI when confronted with this type of gobblygook.

NOT NECESSARY

This idea is not necessary for generative AI to be effective in your organisation. Instead of chasing academic buzzwords, focus on practical implementation:

• Teach people how to communicate effectively with AI (e.g., refining prompts for clarity and relevance).
• Leverage AI training to simplify and improve task efficiency and effectiveness in every role, without worrying about the underlying math.

In short: You don’t need to understand vectors to make generative AI work for your business. That’s just a distraction. What matters is whether your team can use AI effectively to save time, improve quality, increase productivity and drive innovation.

I am confining my remarks here to generative AI, e.g. ChatGPT, Gemini, Claude 3.5 Sonnet (the confounding of which with other forms of AI that have been with us for decades is a source of endless confusion in the advice sprouted from Generative AI systems themselves).

Here's the agenda:

1. Is Thinking And Communicating In Vectors At All Necessary or Helpful?

2. Embedding Generative AI Thinking

3. The Pragmatist Approach

Let’s cut through the academic jargon and examine whether understanding vectors is actually necessary for a CEO or their team to effectively use generative AI in the workplace.

1. Do you need to understand vectors to use generative AI?

No. Generative AI systems, such as ChatGPT, already handle all the complexity of vectors behind the scenes. They translate inputs (text, images, etc.) into vectors to compute results, but users never need to think about this directly. You can prompt an AI to write a business plan, generate a sales pitch, or brainstorm product ideas without ever worrying about how it encodes your input.

2. What does "thinking and communicating in vectors" even mean?

Vectors are mathematical representations of relationships between data points. In AI, this enables concepts like:

• A customer query ("How do I set up this product?") being matched to relevant knowledge base articles.
• Images being clustered by visual similarity.
• Text input ("Give me ideas for growth") being mapped to creative outputs.

For a machine, "thinking in vectors" is useful because it helps process context and similarity. For humans? It’s largely irrelevant unless you're building or training AI systems.

3. What’s the real-world practicality for a CEO or their employees?

As a CEO, your primary concern isn’t learning vectors but ensuring AI is useful and impactful for your business:

• Do employees need to know how AI encodes their input? No. They need to know how to phrase their queries effectively to get good results.
• Will understanding vectors improve workplace productivity? Doubtful. What improves productivity is training employees to use AI tools strategically for tasks like content generation, summarizing information, or making decisions.

4. So, is the claim just hype?

Mostly, yes. The phrase "thinking and communicating in vectors" might sound profound, but it’s not practically relevant for:

• The end-user of AI.
• Leaders implementing AI in their organizations.

It’s a misdirection. Instead of telling employees they need to think about vectors, the focus should be on AI Thinking.

Embedding Generative AI Thinking

"AI Thinking" is the mindset and intuitive skill set that enables employees to instinctively use generative AI as their first resource for solving problems, completing tasks, and overcoming obstacles. It emphasises building confidence, fluency, and measurable impact in guiding AI responses to achieve better organisational outcomes.

To achieve AI thinking, four key implementation steps are required.

1. Measuring ROI on Generative AI Efforts

To ensure "AI Thinking" delivers tangible value, you must integrate generative AI usage into performance metrics at all levels, specifically productivity measures.

Individual-Level Metrics:

1. Evaluating how effectively employees use generative AI to streamline tasks, solve problems, or innovate in their roles. Effectiveness requires systemised training focused on skill acquisition related to specific job roles and outcomes and overall generative AI capabilities.

Examples:

• Time saved on repetitive tasks (e.g., drafting reports, summaries, or plans).
• Quality of outputs compared to traditional methods (e.g., marketing copy effectiveness or customer satisfaction improvements).

2. Using periodic feedback sessions or self-assessments to gauge confidence in AI use and identify training needs (and see #3 below - visualising AI adoption).

Departmental and Functional Metrics:

1. Measuring adoption rates within teams and functions.
2. Assessing productivity gains, such as project completion speeds or cost reductions driven by AI-assisted solutions.

Examples:

• The number of customer service tickets resolved faster using AI-generated responses.
• Marketing campaign turnaround times reduced through AI brainstorming and content creation.

Organisational Metrics:

1. Tracking generative AI’s overall contribution to organisational KPIs like revenue growth, employee satisfaction, or operational efficiency.
2. Identifying and analysing patterns of adoption across departments to pinpoint leaders and laggards.

Examples:

• Percentage of people actively using AI in key deliverables.
• Financial ROI, such as savings or revenue growth attributable to AI-driven improvements.

2. Performance Metrics as a Cultural Lever

Highlighting generative AI usage as a valued skill across the organisation:

• Rewarding employees who consistently and effectively apply generative AI in their roles.
• Recognizing teams or departments driving exceptional results through innovative AI applications.

Example: Establish an internal "AI Innovator" award to celebrate employees who find transformative ways to use AI tools.

3. Visualising AI Adoption

Creating dashboards to provide executives with real-time insights into generative AI's impact:

• Adoption trends across the organisation.
• Productivity and efficiency gains attributable to AI usage.
• Comparative analyses of high- and low-adopting teams to identify best practices or barriers.

4. Fostering AI Leadership

CEOs and leadership teams should champion "AI Thinking" by:

• Encouraging the visible application of AI in their decision-making, e.g. making it the "PhD Level Advisor" at every meeting.
• Sharing success stories of how generative AI has driven measurable results within the organisation.

Why This Matters

1. Accountability for Impact: Measuring the ROI of generative AI ensures that its adoption is not just theoretical or anecdotal but grounded in data and linked to organizational success.
2. Highlighting High Performers: Recognizing and rewarding individuals and teams who excel in AI usage motivates broader adoption and innovation.
3. Driving AI Maturity: With metrics in place, companies can identify areas for improvement, scale successes, and continually refine their "AI Thinking" strategy.

Pragmatic generative AI training focused on roles and outcomes, along with the embedding of AI Thinking as outlined above, will deliver everything possible from current generative AI systems, let alone those coming - which is of immense value for most businesses.

There's no need to think and communicate in vectors.

The Pragmatist Approach

At Simple Academy, we keep things simple and emphasise the importance of focusing on enhancing current job functions rather than becoming overly ambitious with predictions about the future:

1. Practical Application of AI: We advocate using AI to streamline and improve existing tasks and processes. The goal is to leverage AI to make current job functions more efficient rather than attempting to predict future trends or roles.
2. Efficiency Over Speculation: We suggest that businesses should prioritise making their operations faster and easier through AI technology instead of getting lost in theoretical discussions. This practical approach ensures immediate benefits from AI integration.
3. Incremental Improvements: We believe in focusing on incremental improvements that AI can bring to current roles and outcomes, which can significantly impact productivity and effectiveness.
4. Solving Present Problems: We know that by concentrating on current challenges that businesses face, organisations can utilise AI to address these issues directly, leading to tangible results and ROI.
5. Foundation for Future Growth: While the focus is on current applications, we know that embedding AI Thinking sklls now will lay a strong foundation for future growth and adaptation as new opportunities and AI advancements are made.

In summary, businesses should harness AI to enhance their existing operations and solve current problems rather than becoming overly focused on future possibilities or esoteric approaches. This pragmatic path delivers immediate benefits and ROI while preparing for future AI changes as and how they arise.

Human-empowered AI productivity training for your team from the Simple Academy, here.