Ask Savannah
What AI actually changes about leadership — and what it doesn't.
AI will replace a great deal of the analysis and execution work leaders currently spend time on, but not the judgment, accountability, and human trust-building that leadership actually requires. The leaders who struggle won't be replaced by AI directly — they'll be leaders who never developed the judgment AI can't provide.
As a tool for speed and analysis on well-defined problems, not as a substitute for the judgment calls that carry real accountability. The leaders getting real value are using it to compress the easy work, freeing more time for the decisions only they can actually own.
Anything requiring accountability, genuine judgment under ambiguity, or the human trust-building that holds a team together — AI can inform those decisions, but it can't own them, and pretending otherwise just hides where the real judgment gap is. The decision still needs a person who can be held responsible for it.
Use it to generate options and analysis, and keep the actual evaluation and decision as work you do yourself — the risk isn't using the tool, it's outsourcing the thinking that's supposed to happen after the tool's output arrives. Treat its answers as a draft to interrogate, not a conclusion to accept.
AI can meaningfully inform an important decision; it shouldn't make one alone, because it has no accountability for the outcome and no access to the full, often unstated context a real decision requires. Trust it for analysis and pattern recognition, not for judgment calls that carry real consequences.
The person who made the call remains accountable — using a tool to inform a decision doesn't transfer responsibility for it, and treating it as though it does is exactly how bad decisions get made without anyone actually owning them. That accountability question is worth settling explicitly, before it's tested by a real failure.
Address the fear directly and honestly rather than reassuring them it isn't real — name specifically what is and isn't likely to change for their role, and invest visibly in helping them build the skills that remain distinctly human. Vague reassurance reads as avoidance; specific honesty reads as leadership.
Less technical fluency with the tools themselves, more judgment about where AI genuinely adds value versus where it introduces risk, and the discipline to evaluate its output critically rather than deferring to it. The executives falling behind aren't the ones who know fewer tools — they're the ones who've stopped asking hard questions of the tools they use.
Apply the same commercial rigor you'd apply to any other investment — what specifically is the expected return, and is it actually materializing, rather than treating adoption itself as evidence of success. Activity around a new tool is not the same as value created by it.
Focus on judgment about when and how to use it well, not on tracking every new tool or capability — the fundamentals of good decision-making change far more slowly than the tools do. Trying to stay current on everything is a losing game; staying sharp on judgment is not.
Have a question of your own?