The Value of Independent Judgement in an Age of AI
On adaptability, resilience and the discipline of thinking for ourselves
Artificial intelligence has made sophisticated analysis more accessible than at any previous point in history. Information can be synthesised, competing arguments examined and recommendations produced in seconds
But greater access to intelligence does not necessarily produce better judgement.
A persuasive answer may rest on incomplete evidence. A technically sound recommendation may overlook commercial realities. And a decision supported by extensive analysis may still be fundamentally wrong.
This raises a question that extends well beyond technology: when answers become increasingly easy to obtain, what distinguishes those who make good decisions?
For leaders, entrepreneurs and investors, the answer lies partly in the ability to recognise the limits of what we know. It also requires the discipline to challenge established assumptions, remain open to contrary evidence and distinguish between circumstances that demand persistence and those that require change.
But flexibility is not inherently a sign of good judgement.
An organisation that repeatedly changes strategy in response to market sentiment may appear adaptable while lacking a coherent understanding of its purpose. Equally, one that refuses to reconsider an unsuccessful approach may mistake persistence for conviction.
The distinction lies in understanding what should remain constant and what must evolve.
Consider a company developing technology to improve the efficiency of electricity networks. Its original commercial strategy may depend on utilities adopting the technology directly. Over time, customer discussions might reveal that procurement cycles are longer than anticipated, while industrial energy users demonstrate a more immediate willingness to pay.
The underlying problem has not disappeared. Nor has the technology necessarily lost its value. What has changed is the evidence concerning its most viable route to market.
An adaptable founder may revise the commercial strategy while preserving the company’s fundamental ambition.
From an investment perspective, this distinction matters.
A founder’s willingness to pivot is frequently interpreted as evidence of adaptability. Yet a pivot is not necessarily a positive development. It may reflect a thoughtful response to new information, or it may expose weaknesses in the original understanding of the market.
The more revealing question is not whether the founder changed direction, but whether the reasoning behind that change was sound.
Conviction without adaptability can become rigidity. Adaptability without conviction can become drift.
Good judgement requires the ability to recognise the difference.
What McKinsey’s Research Tells Us About Adaptability
McKinsey’s 2021 article, Future proof: Solving the adaptability paradox for the long term, provides a useful framework for understanding why adapting to uncertainty can be difficult, even for experienced leaders.
The research examines how individuals and organisations can develop adaptability as a sustained capability rather than a temporary response to disruption.
Its five recommended actions encompass personal wellbeing, clarity of purpose, an adaptive mindset, meaningful human connections and an environment in which learning is psychologically safe.
An important implication is that adaptability does not emerge simply from exposure to change. It requires the conditions and habits that allow people to respond thoughtfully rather than defensively.
This distinction deserves particular attention.
Experience, while valuable, can reinforce familiar patterns of thinking. Under pressure, leaders may return to established methods precisely when circumstances demand a different response.
A strong sense of purpose can provide direction, but it must be accompanied by a willingness to reconsider the assumptions underlying a particular strategy.
Similarly, psychological safety matters because individuals are less likely to challenge prevailing views when disagreement carries professional or personal consequences.
These principles extend beyond organisational culture. They influence the quality of investment decisions, entrepreneurial judgement and strategic leadership.
Yet the growing influence of artificial intelligence introduces another dimension to the discussion.
What happens when the information informing our decisions is increasingly generated, organised and interpreted by systems whose outputs can appear authoritative even when their underlying assumptions are uncertain?
The Paradox of Intelligence Without Judgement
Artificial intelligence can be an extraordinary analytical resource.
It can help identify patterns across large datasets, compare competing scenarios and make specialised knowledge more accessible. For entrepreneurs and smaller organisations, capabilities once dependent on substantial analytical resources are increasingly within reach.
But the speed at which information is produced can create a misleading impression of certainty.
An AI-generated assessment of an emerging technology might accurately describe its technical potential while underestimating the difficulty of commercial adoption. A market analysis may identify substantial demand without sufficiently accounting for procurement constraints, regulatory uncertainty or the economics of implementation.
The problem is not necessarily that the information is incorrect.
It is that the analysis may be incomplete in ways that are difficult to recognise without independent expertise.
Research published in 2025 by researchers from Microsoft Research and Carnegie Mellon University offers an interesting perspective.
In a study involving 319 knowledge workers and 936 reported examples of generative AI use, researchers found that greater confidence in AI was associated with lower self-reported critical-thinking effort. Conversely, greater confidence in one’s own ability to perform a task was associated with more critical-thinking engagement.
The study does not establish that AI causes a decline in critical-thinking ability. It does, however, raise an important question about how reliance on technology may influence the effort people devote to evaluating information.
The researchers also observed that critical-thinking activities were shifting towards verifying information, integrating AI-generated responses and overseeing the work being performed.
This suggests that the nature of human judgement may be changing rather than simply becoming less important.
As technology assumes more analytical tasks, the responsibility to assess the quality, relevance and consequences of its outputs becomes increasingly significant.
The ability to produce an answer is becoming easier to acquire. The ability to recognise whether it is the right answer remains far more demanding.
Why Independent Thinking Matters in Investment Decisions
Investment provides a particularly useful setting in which to examine this distinction.
An investment decision is rarely made with complete information. Financial projections depend on assumptions. Market opportunities are shaped by external conditions. Technological developments may advance faster than regulation, infrastructure or customer adoption.
This is especially apparent in climate and industrial technologies, where technical feasibility and commercial viability do not always progress at the same pace.
A technology may offer substantial environmental benefits without yet having a sustainable business model. A founder may possess exceptional technical expertise while underestimating the complexity of bringing a product to market.
The challenge is to distinguish between genuine long-term potential and a compelling narrative unsupported by sufficient evidence.
AI can contribute meaningfully to this assessment. It can help examine industry trends, identify competitors, compare business models and test financial assumptions.
But it cannot remove the fundamental uncertainty of investing.
Nor should the sophistication of an analytical model be confused with the strength of the evidence on which it depends.
An investor must still determine which assumptions are credible, which risks are acceptable and what evidence would justify changing an initial view.
This is where independent judgement becomes essential.
A useful discipline is to ask not only why an opportunity appears attractive, but what would need to be true for the investment thesis to fail.
Such an approach does not require excessive scepticism. It requires intellectual honesty.
The objective is not to eliminate uncertainty, which is rarely possible, but to understand it well enough to make a considered decision.
Resilience Requires More Than Persistence
The relationship between adaptability and resilience is equally important.
Resilience is commonly associated with perseverance: the ability to withstand setbacks, recover from disappointment and continue pursuing an objective.
Yet persistence can become counterproductive when it is disconnected from evidence.
An entrepreneur may continue financing an unsuccessful strategy because considerable time and capital have already been invested. An organisation may defend an established business model because changing it would require acknowledging that earlier assumptions were mistaken.
In such circumstances, continuing is not necessarily a demonstration of resilience.
It may simply be a reluctance to confront an uncomfortable conclusion.
Genuine resilience includes the ability to reassess.
Sometimes progress requires greater determination. At other times, it requires the confidence to abandon an approach that no longer serves its intended purpose.
The distinction is not always obvious, particularly when decisions carry emotional, financial or reputational consequences.
This is why resilience should be understood alongside adaptability rather than independently of it.
Resilience enables us to remain engaged with difficult challenges. Adaptability allows us to reconsider how those challenges should be approached.
Independent judgement helps determine which response is appropriate.
Together, they support a form of perseverance that is neither blindly optimistic nor prematurely defeated.
The Intellectual Discipline of Changing One’s Mind
Independent thinking is sometimes mistaken for the willingness to disagree with conventional opinion.
But disagreement alone does not demonstrate intellectual independence.
A contrarian position can be just as poorly examined as a widely accepted one.
The more demanding discipline is to question our own conclusions with the same rigour we apply to those of others.
Professional experience can make this difficult. As expertise develops, so does confidence in familiar methods of analysis. Patterns that have proved useful in the past may gradually become assumptions that are no longer examined.
Yet changing one’s mind in response to stronger evidence is not necessarily a sign of inconsistency.
It can be evidence of intellectual maturity.
For leaders, this requires creating environments in which informed disagreement is valued rather than discouraged.
For investors, it means remaining willing to revisit an investment thesis when circumstances change.
For entrepreneurs, it may involve recognising that a deeply held belief about a product, customer or market is no longer supported by the available evidence.
The central question is simple:
What evidence would persuade me that my current understanding is wrong?
An inability to answer does not necessarily indicate conviction. It may reveal that a conclusion has become insulated from scrutiny.
In an age of AI, this question becomes particularly valuable.
Technology can be used to reinforce an existing position by generating increasingly sophisticated supporting arguments. Equally, it can be used to examine alternative explanations, identify overlooked risks and construct the strongest possible challenge to an initial conclusion.
The difference lies in how deliberately we choose to use it.
Independent Judgement Does Not Mean Thinking Alone
There is a tendency to associate independent thinking with individual intellectual strength.
In practice, good judgement often depends on recognising the limitations of one’s own expertise.
Complex decisions benefit from multiple perspectives.
An investor evaluating an energy technology may require technical insight from engineers, commercial knowledge from industry operators and an understanding of the regulatory environment.
No single perspective captures the entire problem.
The purpose of independent judgement is therefore not to reject external expertise, but to evaluate it thoughtfully.
This applies equally to human advice and AI-generated analysis.
A recommendation should be examined in relation to its evidence, assumptions and relevance to the decision at hand, rather than accepted because of the authority or sophistication of its source.
Collaboration can strengthen independent thinking when it exposes us to perspectives we would not otherwise consider.
The challenge is to remain intellectually open without becoming intellectually dependent.
What Should Remain Human?
As artificial intelligence becomes embedded in research, investment analysis and organisational decision-making, the question of which tasks should remain human will continue to evolve.
Many analytical activities will become faster and more accessible. Some decisions may increasingly be supported, or even executed, by automated systems.
But important distinctions remain.
Determining whether a technology can work is not the same as deciding whether it should be deployed.
Identifying a commercially attractive opportunity is not equivalent to determining whether its risks are acceptable.
Producing a recommendation is different from accepting responsibility for its consequences.
These distinctions involve values, institutional context, accountability and judgement.
They cannot be resolved through computational capability alone.
The most effective relationship between human intelligence and artificial intelligence may therefore be one in which technology expands our capacity to examine possibilities while people retain responsibility for defining objectives, evaluating trade-offs and making consequential decisions.
This is not an argument for preserving human involvement simply because a task has historically been performed by people.
It is an argument for recognising that analytical capability and responsible judgement are related, but not interchangeable.
The Enduring Advantage
The increasing sophistication of artificial intelligence will continue to transform how information is produced and decisions are supported.
For individuals and organisations, the opportunity is considerable. Better tools can expand access to knowledge, improve analytical capacity and make previously complex tasks more manageable.
Yet these advantages will not automatically translate into better decisions.
Adaptability will remain essential because circumstances change and established assumptions can lose their relevance.
Resilience will remain important because meaningful progress involves uncertainty, setbacks and difficult choices.
Independent judgement will matter because neither experience nor technology can eliminate the responsibility to think critically about the conclusions we reach.
The challenge is not to compete with machines in producing information, nor to reject the capabilities they offer.
It is to become more discerning in how we use them.
That means recognising when an answer is incomplete, when evidence challenges an established belief and when a decision requires a different course of action.
In an age when intelligence is becoming increasingly accessible, the enduring advantage may lie in knowing how to exercise judgement.
References
- Brassey, J., De Smet, A., Kothari, A., Lavoie, J., Mugayar-Baldocchi, M., and Zolley, S. (2021). Future proof: Solving the ‘adaptability paradox’ for the long term. McKinsey & Company, 2 August 2021. https://www.mckinsey.com/capabilities/people-and-organization/our-insights/future-proof-solving-the-adaptability-paradox-for-the-long-term
- Lee, H.-P., Sarkar, A., Tankelevitch, L., Drosos, I., Rintel, S., Banks, R., and Wilson, N. (2025). The Impact of Generative AI on Critical Thinking: Self-Reported Reductions in Cognitive Effort and Confidence Effects From a Survey of Knowledge Workers. Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems. https://doi.org/10.1145/3706598.3713778