Balancing AI and Human Expertise

AI will always find you the fastest path. The dangerous part is assuming fastest and right are the same thing. 🧠 After 13 years and 200+ enterprise AI deployments, this is the distinction I keep coming back to in every system I build and every leadership team I work with. AI optimises for efficiency. It does not have access to the relationship history, the political context, the ethical weight, or the lived experience that determine whether the efficient path is actually the right one for this specific situation, with these specific people, right now. That is not a flaw to be engineered away. It is the permanent and irreplaceable role of human judgment. Here is a framework I use when working with AI outputs on high-stakes decisions. ➡️ What context does this decision require that AI does not have access to? ➡️ What would I decide if I had not seen the AI recommendation first? ➡️ Am I using this output to inform my thinking or replace it? The third question is the most important. Using AI to inform your thinking is amplification. Using it to replace your thinking is atrophy. And the line between the two is easier to cross than most people realise. What is one decision in your work where you would never let AI have the final say? #ai #leadership #futureofwork #artificialintelligence #aistrategy #teamhuman #criticalthinking #intellectualatrophy

There is a lot of talk about AI, data, analytics, and algorithms. And they all matter. But for the next several years, human judgment is what will matter most. AI can process 1.7 billion variables. The human mind can deal with four to six. That gives us better data, better options, and more consistency. All of that helps. But judgment is still not quantifiable. It’s processed by the brain in ways we don’t fully understand. And we never will. Judgment shows up in places AI cannot reach: 🔹 How you define the problem 🔹 Which assumptions you challenge 🔹 Which risks you take 🔹 Which goals you choose 🔹 How much ego damage you can stand Good judgment comes from bad experiences. That’s been true for centuries. When something goes wrong, reflect: 🔹 Were the facts wrong? 🔹 Was the reframing wrong? 🔹 Was the weighting of facts wrong? 🔹 Or was it a judgment shaped by bias or risk preference? Use AI to expand your thinking. But don’t outsource judgment. Strengthen it. That’s what leaders are admired for.

Last week, a senior manager presented me with a strategic roadmap during an advisory session. It was polished, grammatically perfect, and filled with current buzzwords. It looked like a fantastic job but my gut feeling gave me a strange feeling I asked one simple question: "𝘞𝘩𝘺 𝘥𝘪𝘥 𝘺𝘰𝘶 𝘱𝘳𝘪𝘰𝘳𝘪𝘵𝘪𝘻𝘦 𝘤𝘩𝘢𝘯𝘯𝘦𝘭 𝘟 𝘰𝘷𝘦𝘳 𝘤𝘩𝘢𝘯𝘯𝘦𝘭 𝘠 𝘪𝘯 𝘘3?" I was not surprised by the reaction. He froze. He couldn't give a proper answer. Why? Because he hadn't made that decision. The algorithm did. He had fallen for the "𝗢𝗿𝗮𝗰𝗹𝗲 𝗠𝘆𝘁𝗵". He treated the AI as a "know-it-all" guru rather than what it actually is: a high-power probabilistic engine. This passive approach is dangerous. When we view AI as an oracle, we stop analyzing and start obeying. We confuse “𝘨𝘰𝘰𝘥 𝘸𝘳𝘪𝘵𝘪𝘯𝘨” with “𝘨𝘰𝘰𝘥 𝘪𝘥𝘦𝘢𝘴 𝘵𝘩𝘢𝘵 𝘐 𝘳𝘦𝘢𝘭𝘭𝘺 𝘶𝘯𝘥𝘦𝘳𝘴𝘵𝘢𝘯𝘥 𝘢𝘯𝘥 𝘐 𝘤𝘢𝘯 𝘸𝘰𝘳𝘬 𝘸𝘪𝘵𝘩”. Here is the uncomfortable reality: LLMs do not "reason" in the human sense; they predict the next most likely word based on patterns. They are designed to sound convincing, not to be factually accurate. If you want to survive the Algorithm Era, you must shift from a passive user to an active driver. Here is how to break the AI toxic dependency:   • 𝗗𝗲𝗺𝗼𝘁𝗲 𝘁𝗵𝗲 𝗔𝗜: Stop treating ChatGPT as a Vice President of Strategy. Treat it as a brilliant but sometimes intoxicated summer intern. It generates volume but YOU provide the judgment.   • 𝗧𝗵𝗲 "𝗝𝗮𝗴𝗴𝗲𝗱 𝗙𝗿𝗼𝗻𝘁𝗶𝗲𝗿" 𝗥𝘂𝗹𝗲: AI excels at creative brainstorming but often fails at simple logical tasks. Never delegate the final decision on high-stakes logic to a black box.   • 𝗜𝗻𝘁𝗲𝗿𝗿𝗼𝗴𝗮𝘁𝗲, 𝗗𝗼𝗻'𝘁 𝗝𝘂𝘀𝘁 𝗔𝘀𝗸: Don't just ask for an answer. Ask the AI to show its work. Force it to reveal its "Chain of Thought" so you can verify the logic, not just the result.   • 𝗢𝘄𝗻 𝘁𝗵𝗲 𝗪𝗵𝘆: If you cannot explain the rationale behind an AI-generated strategy without looking at your notes, you do not have a strategy. You have a hallucination. Let’s be honest: What is the most plausible lie an AI has told you recently that almost slipped into a final report?. I’ll start: AI confidently claimed a competitor had discontinued a specific product line because it seemed "logical." It hadn't. Let me know in the comments. #AIAugmentedProfessional #HybridIntelligence #AiforExecutives #OracleMyth

Over the past few weeks, I’ve had time to reflect — between AI briefing sessions, leading workshops, and teaching at LinkedIn offices in London. One question keeps coming up: “Will I still have a job?” It’s a fair question. The reports, social posts, and videos can be overwhelming — even frightening. But here’s what I’ve come to believe: -- AI isn’t replacing jobs outright. It’s replacing tasks. And that changes everything. In my role as Course Director at the CIM | The Chartered Institute of Marketing — and through my work with marketers, consultants, and business owners — I’ve seen the same shift again and again: Those who are moving forward are reframing their role. They’re not competing with AI — they’re learning how to work alongside it. Here’s how I break down the Human + AI relationship in practical terms: 👉 20% – Spark Ideas Use AI to generate ideas, explore angles, and overcome creative blocks. 👉 40% – Co-Pilot Mode Let AI support your process — drafting, outlining, and iterating while you steer. 👉 60% – Efficiency Booster AI automates and structures routine work so you can focus on higher-value thinking. 👉 80% – Heavy Lifter AI takes on the complexity. You bring the insight, oversight, and strategic direction. 👉 100% – Full AI Execution In some cases, AI completes the task. But the human still sets the brief, guides the tone, and makes the decisions. For Marketers, Consultants, and Business Owners, this is not the time to resist. It’s time to rethink your value, redistribute your time, and reimagine your workflows. Humans won’t be replaced.... But we will be expected to become smarter — by collaborating with AI. Of course, there’s a much bigger discussion to be had. This post is just one thought in that wider conversation. But it’s a good place to start. With Positivity, Imran. #AIinMarketing #MarketingLeadership #CharteredInstituteOfMarketing #FutureOfWork #AIMarketer

If students don’t learn how to think with AI, they’ll let AI think for them. Last Thursday at Shanghai American School, I got to "beam in" to give a keynote presentation on one of the most urgent conversations in education today: How do we integrate AI without losing what makes learning human? Here are the key takeaways from our time together: • Generative AI can amplify learning—or weaken it. Studies show that when students engage critically with AI, they learn more. But when they rely on it to do the work for them, learning declines. The key? Teach students to think with AI, not just use it. • Confidence in AI can lower critical thinking. Research suggests that when people trust AI too much, they question it less. The best educators will teach students how to balance trust and skepticism when using AI tools. • Ethical AI use starts with values. We discussed how every school needs guiding principles for AI integration—beyond just policies. What should we protect? What should we enhance? These questions shape AI’s role in education. We concluded with "Three Ts" for responsible AI use: 1. Talk – Normalize generative AI discussions with students and teachers. I shared my "Generative AI Guidelines Canvas" to support conversations. https://lnkd.in/gyjTkK7d 2. Teach – Build generative AI literacy into the curriculum. I shared Cora Yang and Dalton Flanagan's C.R.E.A.T.E. framework for teaching students to prompt. https://lnkd.in/g-KYt4Uy 3. Try – Teachers should experiment with generative AI tools in meaningful, ethical ways. I shared Darren Coxon's Hattie Bot to let teachers experiment with building lessons that have high effect size. https://lnkd.in/g44gZzA3 This conversation isn’t over—it’s just beginning. Critical thinking isn't optional if machines do the easy thinking for us. Much gratitude to Alan Preis & Scott Williams for crafting such a great experience. Photo Credit Alex McMillan 🙏 P.S. I asked everyone at Shanghai American School: What values should guide our approach to AI in education? What's your answer? #generativeAI #guidelines #teachers #ethics

Reliability, evaluation, and “hallucination anxiety” are where most AI programmes quietly stall. Not because the model is weak. Because the system around it is not built to scale trust. When companies move beyond demos, three hard questions appear: →Can we rely on this output? →Do we know what “good” actually looks like? →How much human oversight is enough? The fix is not better prompting. It is a strategy and operating discipline. 𝐅𝐢𝐫𝐬𝐭: ⁣Define reliability like a product, not a vibe. Every serious AI use case should have a one-page SLO sheet with measurable targets across: →Task success ↳Right-first-time rate and rubric-based acceptance →Factual grounding ↳Evidence coverage and unsupported-claim tracking →Safety and compliance ↳Policy violations and PII leakage →Operational quality ↳Latency, cost per task, escalation to humans Now “good” is no longer opinion. It is observable. 𝐒𝐞𝐜𝐨𝐧𝐝:  evaluation must be continuous, not a one-off demo test. Use a simple loop: 𝐏lan: Define rubrics, datasets, and risk tiers 𝐃⁣o: Run offline evaluations and limited pilots 𝐂heck: Monitor drift and regressions weekly 𝐀ct: Update prompts, data, guardrails, and workflows Support this with an AI test pyramid: →Unit checks for prompts and tool behaviour →Scenario tests for real edge failures →Regression benchmarks to prevent backsliding →Live monitoring in production Add statistical control charts, and you can detect silent degradation before users do. 𝐓𝐡𝐢𝐫𝐝: reduce hallucinations by design. →Run a short failure-mode workshop and engineer controls: →Require retrieval or evidence before answering →Allow safe abstention instead of confident guessing →Add claim checking and tool validation →Use structured intake and clarifying flows You are not asking the model to behave. You are designing a system that expects failure and contains it. 𝐅𝐨𝐮𝐫𝐭𝐡: make human-in-the-loop affordable. Tier risk: →Low risk: Light sampling →Medium risk: Triggered review →High risk: Mandatory approval Escalate only when signals demand it: low confidence, missing evidence, policy flags, or novelty spikes. Review becomes targeted, fast, and a source of improvement data. 𝐅𝐢𝐧𝐚𝐥𝐥𝐲: Operate it like a capability. Track outcomes, risk, delivery speed, and cost on a single dashboard. Hold a short weekly reliability stand-up focused on regressions, failure modes, and ownership. What you end up with is simple: ↳Use case catalogue with risk tiers ↳Clear SLOs and error budgets ↳Continuous evaluation harness ↳Built-in controls ↳Targeted human review ↳Reliability cadence AI does not scale on intelligence alone. It scales on measurable trust. ♻️ Share if you found thisuseful. ➕ Follow (Jyothish Nair) for reflections on AI, change, and human-centred AI #AI #AIReliability #TrustAtScale #OperationalExcellence

A software engineer can ask AI to solve a complex programming problem in a few precise words, drawing on years of technical knowledge. But someone without programming experience? They might need paragraphs just to explain what they want, often missing crucial technical context and requirements that would be second nature to a developer. This highlights a key truth: AI is an incredible force multiplier for those who understand the fundamentals. It's not just about giving commands – it's about knowing which problems need solving and how to frame them effectively. Yes, AI democratizes access to programming capabilities. But it simultaneously increases the value of deep technical expertise. The most powerful combination? Domain knowledge + AI fluency.

AI products like Cursor, Bolt and Replit are shattering growth records not because they're "AI agents". Or because they've got impossibly small teams (although that's cool to see 👀). It's because they've mastered the user experience around AI, somehow balancing pro-like capabilities with B2C-like UI. This is product-led growth on steroids. Yaakov Carno tried the most viral AI products he could get his hands on. Here are the surprising patterns he found: (Don't miss the full breakdown in today's bonus Growth Unhinged: https://lnkd.in/ehk3rUTa) 1. Their AI doesn't feel like a black box. Pro-tips from the best: - Show step-by-step visibility into AI processes - Let users ask, “Why did AI do that?” - Use visual explanations to build trust. 2. Users don’t need better AI—they need better ways to talk to it. Pro-tips from the best: - Offer pre-built prompt templates to guide users. - Provide multiple interaction modes (guided, manual, hybrid). - Let AI suggest better inputs ("enhance prompt") before executing an action. 3. The AI works with you, not just for you. Pro-tips from the best: - Design AI tools to be interactive, not just output-driven. - Provide different modes for different types of collaboration. - Let users refine and iterate on AI results easily. 4. Let users see (& edit) the outcome before it's irreversible. Pro-tips from the best: - Allow users to test AI features before full commitment (many let you use it without even creating an account). - Provide preview or undo options before executing AI changes. - Offer exploratory onboarding experiences to build trust. 5. The AI weaves into your workflow, it doesn't interrupt it. Pro-tips from the best: - Provide simple accept/reject mechanisms for AI suggestions. - Design seamless transitions between AI interactions. - Prioritize the user’s context to avoid workflow disruptions. -- The TL;DR: Having "AI" isn’t the differentiator anymore—great UX is. Pardon the Sunday interruption & hope you enjoyed this post as much as I did 🙏 #ai #genai #ux #plg

Some thoughts on how we integrate AI into education: We first need to start by recognizing which skills are becoming more valuable and designing new ways to teach them. We all remember the effort it takes to write a paper—revising, structuring arguments, and refining our points. With AI, everyone will have a writing co-pilot to handle the mechanics, making the process more efficient. So, what if we redirected that effort into helping students develop higher-order skills like critical thinking, prompt design, and iterative analysis? A thought experiment: Imagine an assignment where students submit not just their essays but also the prompts they used to get AI-generated critiques. Their task wouldn’t be just to write and submit—it would be to argue, analyze, refine, and iterate. In less time than it takes to write a traditional paper, students could engage in deeper intellectual exercises—interrogating their own arguments, considering counterpoints, and strengthening their reasoning. For teachers, AI can streamline grading while amplifying feedback—providing broad insights that help shape targeted, meaningful commentary. This means students receive richer, more personalized guidance, making learning more interactive and impactful.

When I work with companies and governments on AI, the first question I get them to ask is WHY. Why do you want this system? Why this system and not a non-AI one? Why are we seeking to develop even more autonomous AI? Surprisingly, many times it's the fundamental questions that are bypassed all together. The most important problem regarding so-called "AI agents" is the same as their most "attractive" feature: "The more autonomous an AI system is, the more we cede human control." When a system acts independently and with access to multiple systems, applications and platforms, "it is likely to perform actions we didn’t intend, such as manipulating files, impersonating users, or making unauthorized transactions. The very feature being sold—reduced human oversight—is the primary vulnerability." Already my phone is doing lots of things that I don't want it to do. I don't want it to collect much of the data it's collecting; I don't want it to send much of the data it's sending; I don't want to need to use my face to unlock it, etc. If part of what it means to have a good life is to have control over your own life, to have self-governance, or what philosophers call autonomy, then giving up control to AI by definition is worsening our lives, lessening our chances of having a good life. Instead of trying to build decision-makers, we should create systems that remain tools, "assistants rather than replacements. Human judgment, with all its imperfections, remains the essential component in ensuring that these systems serve rather than subvert our interests." Article by Margaret Mitchell, Dr. Sasha Luccioni, and Avijit Ghosh, PhD. #AIEthics https://lnkd.in/enfFT2mi