The New Renaissance Man
Imagine a senior partner in a law firm, a retired trader, a former diplomat and a machine-learning engineer sitting around the same table in 2028. The engineer understands the model architecture. The lawyer understands liability. The trader understands incentives, timing and second-order effects. The diplomat understands signalling, institutions and human vanity. The strange fact is that the most useful person in the room may not be the one with the deepest technical knowledge of artificial intelligence, but the one able to recognise which parts of the problem are technical, which are legal, which are strategic, which are psychological and which are simply human. The machine has changed the meeting, but not in the way many expected. It has made knowledge more available while making judgement more exposed.¹
For much of the twentieth century, the safest advice to an ambitious young person was to specialise. Become a surgeon, a tax lawyer, a chip designer, a quantitative analyst, a molecular biologist. Modern civilisation rewarded depth because knowledge was scarce, training was expensive and expertise took years to acquire. The specialist possessed something others could not easily obtain: command of a difficult field whose language, methods and assumptions were inaccessible to outsiders.²
Artificial intelligence does not abolish that advantage, but it weakens its monopoly. A capable model can now provide a competent introduction to almost any field within seconds. It can summarise a legal doctrine, explain a statistical technique, compare historical analogies, translate technical jargon and identify the main schools of thought within a debate. It cannot replace the best specialist, and the pretence that it can is one of the lazier errors of the moment. But it does reduce the distance between ignorance and useful participation. The threshold for entering a field is falling.³
That changes the economics of human value. When information is scarce, the person who possesses it commands a premium. When information becomes abundant, value migrates towards those who can ask better questions, detect weak assumptions, compare incompatible frameworks and decide which knowledge matters in a particular situation. The scarce faculty becomes not memory, nor even expertise alone, but judgement under uncertainty.
This is why the return of the Renaissance man should not be understood nostalgically. No modern Leonardo can master the accumulated knowledge of physics, biology, law, finance, medicine, engineering and political economy. The body of knowledge is now too large. Yet the Renaissance ideal was never merely encyclopaedic accumulation. Its deeper significance lay in the refusal to accept that reality itself was divided into the categories later imposed by universities and professions. Leonardo moved between anatomy, hydraulics, optics and painting because the world he studied was continuous.⁴
The modern synthesist returns for a different reason. He does not know everything. He knows how to move. Artificial intelligence supplies bridges across fields that were previously separated by years of prerequisite knowledge. The human advantage increasingly lies in recognising which bridge is worth crossing.
Several older thinkers anticipated parts of this problem without seeing its present form. Michael Polanyi distinguished explicit knowledge from tacit knowledge, reminding us that much of what humans know cannot be fully reduced to rules or propositions.⁵ Herbert Simon understood that intelligence operates under bounded rationality, where decisions must be made without complete information.⁶ Friedrich Hayek argued that knowledge is dispersed through society and cannot be centrally possessed.⁷ Daniel Kahneman later showed how judgement is shaped by cognitive bias, intuition and context rather than information alone.⁸ Together, they point towards a conclusion now made sharper by AI: the most valuable human capacities are often those hardest to formalise.
The error made by many discussions of artificial intelligence is to treat knowledge work as if it consisted mainly of producing correct answers. Much of it does not. It consists of framing the problem, deciding what kind of answer would count as useful, identifying the incentives of the people asking, recognising when a technically correct answer is politically impossible, and sensing when the official question conceals the real one. These are not mystical abilities. They are forms of accumulated judgement, often acquired through decades of experience across different domains.
The new Renaissance man is therefore not a dilettante. Nor is he merely a generalist in the shallow sense of someone who knows a little about many things. He is a synthesist: someone capable of using artificial intelligence to range widely while bringing to that range the one thing the machine does not possess—lived, historical, institutional and moral judgement.
Footnotes
- Herbert A. Simon, The Sciences of the Artificial (1969); Michael Polanyi, Personal Knowledge (1958).
- Max Weber, “Science as a Vocation” (1917); Adam Smith, The Wealth of Nations (1776).
- OpenAI, GPT-4 Technical Report (2023); Anthropic, Claude 4 System Card (2025).
- Leonardo da Vinci, Notebooks; Walter Isaacson, Leonardo da Vinci (2017).
- Michael Polanyi, Personal Knowledge (1958); The Tacit Dimension (1966).
- Herbert A. Simon, Administrative Behavior (1947); Models of Man (1957).
- Friedrich Hayek, “The Use of Knowledge in Society” (1945).
- Daniel Kahneman, Thinking, Fast and Slow (2011).
Convergence Essay #10 – Part 2
If this diagnosis is correct, the labour market is about to reward a different kind of intellectual architecture. For nearly two centuries, industrial society steadily decomposed complex work into specialised tasks. Factories divided production into discrete operations; universities divided knowledge into departments; corporations divided responsibility into functions. The arrangement generated extraordinary gains in productivity because it mirrored the central economic reality of the age: information was expensive, expertise was scarce and communication across domains was slow. The specialist became the indispensable unit of progress.⁹
Artificial intelligence changes each of those assumptions simultaneously. Information becomes abundant, explanation becomes inexpensive and translation between disciplines approaches real time. What remains stubbornly resistant to automation is deciding which explanations deserve confidence, which trade-offs matter, and which apparently unrelated developments belong to the same emerging pattern. In other words, AI commoditises answers while increasing the value of questions.
This helps explain a development that initially appears paradoxical. Frontier AI laboratories, venture capital firms and technology founders increasingly speak less about technical brilliance alone and more about judgement, taste, curiosity and breadth.¹⁰ These qualities sound suspiciously like the virtues associated with the humanities rather than the stereotypical engineer. The reason is straightforward. As technical capability becomes increasingly available through intelligent systems, comparative advantage shifts towards recognising opportunities that no benchmark can measure.
History offers several precedents. Warren Buffett rarely attributes his success to superior mathematics. Charlie Munger repeatedly argued that the most durable decisions emerged from what he called a “latticework of mental models”, drawing simultaneously upon psychology, economics, biology, engineering and history.¹¹ John Boyd revolutionised military strategy not by inventing better aircraft but by synthesising ideas from thermodynamics, evolutionary theory and decision science into the OODA loop. None of these figures succeeded because they possessed the deepest expertise in one discipline. They became unusually effective because they recognised recurring structures across many.
Artificial intelligence magnifies that advantage. A machine can retrieve an economic paper in seconds. It cannot yet know whether the assumptions behind the model resemble those that failed during the Asian Financial Crisis, whether an analogy from Roman frontier policy illuminates contemporary geopolitics, or whether a psychological bias explains why an apparently elegant solution repeatedly disappoints in practice. Such judgements emerge from experience accumulated across decades rather than from information acquired in minutes. Michael Polanyi’s tacit knowledge becomes not less valuable in the age of AI, but more.¹²
This is also why I suspect universities may gradually rediscover the importance of broad intellectual formation. The nineteenth century required specialists because knowledge was expanding faster than individuals could absorb it. The twenty-first century still requires specialists to advance the frontier, but it increasingly requires synthesists to navigate the frontier. Degrees that cultivate historical reasoning, institutional analysis, philosophy, literature and political judgement may become unexpectedly valuable, not because artificial intelligence struggles with technical subjects, but because it makes technical competence more widely available while leaving ambiguity stubbornly human.
The irony is striking. For decades many regarded the humanities as economically vulnerable to technological change. Artificial intelligence may instead increase the premium on precisely those disciplines concerned with interpretation, judgement, ethics, history and competing perspectives. Not because they resist automation entirely, but because they train habits of mind that become more valuable as information itself becomes abundant.
Perhaps that is the true meaning of the New Renaissance Man. He is not the person who knows the most. Nor is he the person with the highest benchmark score or the deepest technical specialism. He is the individual capable of moving confidently between economics and biology, between engineering and history, between philosophy and geopolitics, recognising that reality itself has never respected the boundaries through which universities organised knowledge. Artificial intelligence does not diminish such a person. It amplifies him.
If future historians seek the defining human adaptation to the age of artificial intelligence, they may not begin with machines at all. They may begin with the unexpected return of a type of thinker who has appeared only occasionally throughout history: someone whose greatest strength lay not in possessing more knowledge than anyone else, but in perceiving relationships that almost everyone else had overlooked.
Footnotes (Part 2)
- Alfred Chandler, The Visible Hand (1977); Herbert A. Simon, Organizations (with James G. March, 1958).
- Dario Amodei, essays and interviews on AI and the future of work; Sam Altman, essays and interviews on AI capability and human judgement.
- Charlie Munger, Poor Charlie’s Almanack; Warren Buffett, Berkshire Hathaway shareholder letters; John Boyd, OODA Loop briefings.
- Michael Polanyi, The Tacit Dimension (1966); David Epstein, Range (2019).