As artificial intelligence becomes more powerful, perhaps our greatest challenge isn’t technological at all. Inspired by Nick Bostrom’s Superintelligence, I reflect on wisdom, humility, human values and our shared responsibility for the future — and what these mean in our everyday lives.
Reflections on Nick Bostrom, humility, and what it means to be fully human in an increasingly intelligent world.
I’ve recently been reading Superintelligence: Paths, Dangers, Strategies, published in 2014 by philosopher Nick Bostrom. It is a demanding book, examining the possibility that humanity might one day create artificial intelligence vastly exceeding our own capabilities — and the extraordinary opportunities and dangers that could follow.
More than a decade later, with artificial intelligence becoming part of everyday life, Bostrom’s questions seem increasingly relevant.
What particularly struck me, however, wasn’t his examination of technological possibilities. It was where he ended up.
After exploring the profound risks of superintelligence, strategies for managing technological development, and the dangers of competition between those developing powerful AI, Bostrom concludes with something fundamentally human.
We need wisdom, sound judgement, cooperation, humility and a sense of responsibility for one another.
It seems an extraordinary place to arrive after such an intellectually formidable examination of technology.
And perhaps that’s precisely the point.
Intelligence is not wisdom
One of the most important distinctions emerging from my reading is that intelligence and wisdom are not necessarily the same thing.
Intelligence can help us solve problems, recognise patterns, acquire knowledge and develop extraordinary capabilities.
Wisdom involves something different. It asks whether those capabilities should be exercised, for whose benefit, with what consequences, and according to whose values.
Humanity has faced versions of these questions throughout history. Scientific discoveries, industrialisation, medical advances and communications technologies have repeatedly changed the opportunities available to us — and the risks we face.
What distinguishes artificial intelligence may be the speed and scale of its development, together with its potential to influence the very processes through which we make decisions and create new knowledge.
Bostrom is particularly concerned about a future in which machine intelligence becomes substantially more capable than human intelligence. That remains a possibility, not an established destination. Today’s AI systems are not the superintelligence he examines.
Nevertheless, their development already makes some of his underlying questions remarkably practical.
Are we developing the human, social and institutional capabilities necessary to manage increasingly powerful technologies?
And are we giving those questions sufficient attention before the next technological advance arrives?
What makes this particularly urgent is the changing assessment of how soon much more powerful AI might arrive. Some leading researchers now consider systems with capabilities matching or exceeding human expertise across many fields possible within the next few years. Others remain considerably more cautious. While there is no agreed timetable for artificial general intelligence, let alone superintelligence, the possibility of such rapid progress challenges us to begin preparing now rather than leaving these questions to some distant future.
The challenge belongs to humanity
Bostrom’s strategic perspective considers how technological development might be directed towards beneficial outcomes. He examines the importance of developing protective capabilities, understanding risks, avoiding dangerous competition and encouraging cooperation.
His final chapter, Crunch Time, goes further. He asks what humanity should prioritise when the stakes are enormous, knowledge is incomplete and even well-intentioned actions might produce unexpected consequences.
These are not simply engineering questions.
They involve governments, businesses, researchers, communities and individuals. They raise questions about power, accountability, equality and the distribution of benefits.
And they bring us face to face with a familiar problem in human history.
We don’t all agree.
We have different beliefs, values, experiences, ambitions and interests. Some people see AI as an extraordinary opportunity for human advancement. Others regard it with deep suspicion. Many are somewhere between, recognising its usefulness while remaining concerned about its consequences.
These differences needn’t be obstacles to progress. They can be essential contributions to understanding it.
Indeed, I wonder whether one of the greatest dangers would be allowing the future of AI to be determined primarily by those with the technical capability or financial resources to develop it.
Technological expertise is indispensable. But it does not automatically confer moral authority.
Whose future are we creating? Who participates in the decisions? Who benefits, and who bears the risks?
Those questions belong to all of us.
From the global to the personal
Yet there is a danger in discussing artificial intelligence only at the level of global strategy and existential risk.
The issues can become so enormous that ordinary individuals feel powerless. We may conclude that these matters are best left to scientists, governments and technology companies.
But the influence of technology is already reaching into our workplaces, homes, relationships and daily decisions.
That is where the wider philosophical questions become personal.
In my work with digital inclusion, particularly supporting older people, I encounter individuals trying to remain independent and confident in a world that increasingly expects them to use digital services.
Sometimes the immediate task is straightforward: understanding a smartphone, navigating an online service or recognising a suspicious message.
But beneath these practical matters lies something more significant.
How do we help people retain their dignity, independence, confidence and control?
I see a connection here with Bostrom’s wider concerns.
At the global level, he asks how humanity might remain in control of technologies potentially more intelligent than ourselves.
At the personal level, we face a related, though different, challenge: how can individuals exercise meaningful control over technologies they may never fully understand?
The answer cannot simply be to make everyone an enthusiastic technology user.
Someone who decides, after understanding its possibilities and limitations, not to use a particular AI system may be exercising just as much good judgement as someone who chooses to embrace it.
Our responsibility is to support informed choice, not prescribe technological enthusiasm.
Humility: an essential human capability
Recently, I’ve found myself reflecting particularly on humility.
It is a quality I value in my own spiritual practice, and one that seems increasingly important as I explore artificial intelligence.
Humility is not a lack of confidence or an unwillingness to express an opinion. Nor does it require us to diminish our achievements or experience.
Rather, I see it as the willingness to recognise the limits of our knowledge, acknowledge uncertainty and remain open to learning from others.
That includes people whose views differ substantially from our own.
A recent conversation with someone who was sceptical about AI reminded me how important this can be. Rather than trying to overcome their objections, the more useful starting point was to listen and understand them.
Their concerns arose from experiences and responsibilities different from mine.
Those experiences matter.
Perhaps this is particularly important in the AI age, when access to vast quantities of information can create the impression that we understand far more than we actually do.
An AI system may produce a convincing explanation in seconds. That does not mean the explanation is correct, complete or appropriate to the circumstances.
And our own confidence in using these systems should not be confused with wisdom about their consequences.
Humility offers a necessary counterbalance.
It encourages us to question, verify, reconsider and occasionally admit that we simply don’t know.
Becoming a full-stack human
I’ve also been attracted recently to the idea of the full-stack human in the AI age.
Borrowed from the language of technology, where a full-stack developer understands several layers of a computing system, the expression suggests something interesting about human development.
Perhaps our response to AI should involve strengthening the full range of capabilities that make us human.
Not merely acquiring technical skills, but developing critical judgement, emotional intelligence, ethical awareness, creativity, curiosity, empathy and the capacity to work constructively with others.
And, importantly, humility.
We don’t need to compete with machines in every aspect of intelligence. Nor should we surrender our judgement simply because machines can perform certain tasks more quickly or effectively.
Instead, we can ask how these technologies might extend our capabilities while preserving the qualities and relationships that give human life meaning.
This isn’t an argument that humans possess some automatic moral superiority over intelligent machines. It is a recognition that our values, responsibilities and relationships must remain central to the decisions we make about using them.
Being a full-stack human, in this sense, means recognising that technological competence is only one part of living well.
The future is not predetermined
There is an unsettling aspect to Bostrom’s work.
The possibility that humanity might create systems it cannot adequately control is sufficiently serious to deserve careful consideration.
But acknowledging a possibility is not the same as predicting an inevitable outcome.
We should be equally cautious about unquestioning technological optimism and the assumption that catastrophe is unavoidable.
Neither provides a satisfactory foundation for wise action.
The future will be shaped by discoveries and developments we cannot presently anticipate, but also by human decisions: how we design systems, distribute their benefits, manage their risks, establish safeguards and respond to those who may be disadvantaged.
There are no guarantees that we will make those decisions well.
Yet uncertainty need not lead to helplessness.
It can encourage curiosity, preparation, cooperation and a willingness to change course when new evidence emerges.
And perhaps it can help us recognise that responsibility is shared, although our power to influence events is not equal.
The decisions of major technology companies and governments will have consequences on a vastly different scale from those of an individual. That makes institutional accountability essential.
But it doesn’t make individual responsibility irrelevant.
Each of us influences others through our choices, conversations, relationships and work.
Bringing wisdom into everyday life
Where does this leave me?
I’m still learning, and increasingly conscious of how much there is to understand.
Reading Bostrom hasn’t provided a neat set of answers. Instead, it has helped me connect some very large questions about humanity’s future with much smaller questions about my own conduct.
Am I listening carefully enough to people who disagree with me?
Am I distinguishing what I know from what I merely believe?
Am I helping others develop their own judgement rather than encouraging dependence on mine — or on an AI system?
Am I contributing, within my own circumstances, to a future in which technology serves human wellbeing rather than simply advancing because it can?
These are questions I can actually do something about.
They also remind me that wisdom is not something we acquire once and possess forever. It requires continuing attention, reflection, experience and the willingness to reconsider our assumptions.
Perhaps that is what humility makes possible.
Bostrom closes Superintelligence by appealing not to technological brilliance alone, but to humanity’s capacity for decency, common sense and responsible action.
That seems a fitting conclusion to a book about the possibility of intelligence greater than our own.
It also feels like a beginning.
For me, the fundamental challenge of the AI age is not simply whether we can create machines capable of extraordinary intelligence.
It is whether we can develop the wisdom, humility and collective responsibility to use those capabilities well.
The measure of our progress may ultimately be not how intelligent our machines become, but how wisely and compassionately we learn to live alongside them.
And that is a conversation in which everyone deserves a voice.
These reflections were prompted by reading the concluding chapters of Nick Bostrom’s Superintelligence: Paths, Dangers, Strategies (Oxford University Press, 2014), together with my own experiences of technology, digital inclusion and personal learning. They are offered as an invitation to discussion rather than a claim to have definitive answers.