AI Is Changing How We Lead

Leadership
05 October, 2026

The more abundant answers become, the more leadership value moves towards judgment and context.

A few weeks ago, I went through a pre-read that had clearly been drafted with AI. It was well argued and covered all bases. I found myself nodding along, mildly impressed. Only in the meeting room the next day did I realise I hadn’t formed my own view on the topic. And neither had anyone else on the team. We had all read the same smooth narrative and arrived at the same smooth agreement.

That realisation was a jolt. Not that there was anything wrong with the pre-read! But it reminded me how AI is changing our thinking and roles as leaders. Even as AI gives us access to abundant knowledge, it could also make us less perceptive and discerning. That is a drawback we cannot afford, especially when so much of leadership is about creating context, challenging consensus and deciding what deserves attention. An article recently published in the Harvard Business Review highlights what’s at stake:

As organizations gain more “intelligence,” their leaders are being trained out of the very capacity that creates competitive advantage – original judgment. By original judgment we mean the uniquely human capacity to see beyond the dominant narrative and make choices that represent your values.

So, this week, let’s explore how AI is sparking a leadership evolution. What does it mean to lead well at a time when machines can increasingly “think” for us?

Research from the McKinsey Global Institute suggests that as we embrace genuine human-AI skill partnerships, “CEOs and other C-suite leaders will not always be the smartest people in the room”. When answers become instant and plentiful, the prized resource moves from expertise to judgement. A leader’s value increasingly comes from crafting the question, disrupting convergence, making trade-offs, setting guardrails, and realising when the answer in front of them misses the point.

As leaders, we cannot simply become consumers of AI output when one of our core responsibilities is to remain responsible for its direction.

What to delegate vs. what to keep

As AI’s scope grows, we must decide what work to hand off to machines and what should remain in the leader’s domain. C-suite coach David Lancefield offers a useful frame:

  • Leader-owned work that needs your judgement: high-stakes calls where AI can’t understand the history, politics, constraints, and consequences
  • Leader-shaped, AI-assisted work that needs your voice: AI can help structure the message but you need to author its meaning
  • AI-enabled, leader-reviewed work that needs your presence: AI does the heavy lifting and you stay close enough to notice what’s left out
  • AI-handled work that doesn’t need your skills: AI takes care of specific, low-stakes tasks, thus protecting your leadership capacity

As AI expands the options, what is your role?

One of the biggest changes is the sheer volume of options now at our disposal. Historically, a leadership team would explore a handful of strategic alternatives because that’s all they had the time and mental bandwidth for. Our toolkits were shaped by these constraints: from SWOT to Porter’s Five Forces, each framework was built around how much we could squeeze into a day’s session.

Now with AI relaxing the cognitive limits around strategy, we can generate and evaluate much larger option sets. In this scenario, leaders add value by:

  • framing the right question
  • setting the criteria and guardrails
  • identifying which options merit deeper exploration
  • recognising when AI is solving the wrong problem
  • spotting weak reasoning behind polished language
  • pushing beyond conventional thinking

Without the above, any solutions AI provides are likely to be mediocre, not to mention generic, since other teams using a broadly similar prompt and model will arrive at a broadly similar answer.

Another interesting possibility is using AI to think with us – instead of for us. Microsoft Research, which has been exploring this possibility, suggests that AI may work better as a “tool for thought”, used to surface tensions, uncertainties, blind spots, and divergent viewpoints – rather than merely producing finished work.

Adapt your leadership for the AI age

Consider the following recommendations as we enter this new chapter of leadership:

1. Decide the allocation out loud.

Before starting any task, explicitly state the level of involvement from you and AI. Left unsaid, this allocation defaults to whatever is habitual or convenient, meaning either everything goes to AI first because it’s fastest – or nothing does because you aren’t familiar enough with AI yet.

2. Let your thinking take shape before prompting.

For high-stakes decisions that fall under Lancefield’s “leader-owned work”, write down your own diagnosis before asking AI for one. What do I think is happening? What am I assuming? What would change my mind? Then bring AI into the picture. This sequence helps preserve independent thinking and stops AI’s framing from becoming your framing – and by extension, the team’s framing.

3. Embed debate into decision-making.

Guard against AI-manufactured groupthink by incorporating dissent into team processes. Use AI to pressure-test and challenge proposals. At the same time, recognise the limits of synthetic dissent, which relies on the same models and information that underlie your option set. Human dissent is just as vital. Assign a rotating team role to argue against the dominant view, especially when the answer looks obvious. Be intentional about recruiting diverse perspectives.

4. Protect your leadership voice.

Shaping values, inspiring people, fostering trust, and navigating uncertainty – this is deeply human work that belongs to leaders. While AI can help you structure communication around these tasks, it’s vital to keep your authentic voice, especially during key moments like charting a new course or admitting an error. Your words should sound like your own, not like an AI template. Try reading your drafts aloud before finalising them – the quickest way to catch inauthentic language.

5. Stay close to lived reality.

Summaries are great, but aggregation can smooth away certain signals. Maintaining unscripted contact with people across the business remains invaluable. Drop by for a chat with frontline teams. Listen in on a customer call. Go on a spontaneous walk-through. This isn’t about nostalgia – it’s about noticing what the AI-generated answer may be missing.

6. Support structured curiosity.

The HBR article mentioned above offers practical advice for leaders to strengthen their “original judgement”. Collect diverse data, especially from the edges – outliers, weak signals, close calls. Resist the urge to arrive at certainty as quickly as possible. Let the ambiguity and discomfort play out. What are they telling you?

7. Model the AI behaviour you want to see.

Beyond policy documents, teams take their AI cues from leaders. Use AI visibly and scrutinise its answers. Admit when it misled you. Set the standard for what “good” looks like, instead of leaving people to guess whether caution or speed will be rewarded.

8. Lead through redesign.

Microsoft’s latest Work Trend Index sounds a clarion call: leaders must be willing to rearchitect their business around AI. The task at hand is to redesign systems and processes to capture the value AI offers. Importantly, the report identifies increased human agency as a key source of this value: as AI and agents take on more execution, people gain room to direct work, make decisions, and own outcomes.

AI is making some sources of leadership value less important while bringing others into sharper focus. Our role is less about knowing all the facts and producing a competent analysis – and more about deciding what to delegate, where to pay attention, which assumptions to challenge, and when to push back against a confident-sounding answer.

AI can expand the knowledge available to us, but it cannot decide what deserves our attention, what trade-offs to make, or what we are prepared to take responsibility for.

As answers become easier to generate, will judgement become the defining leadership advantage?

X

Join the 8AM conversation

Loading