The fourth decade of McKinsey’s journey, from 1956 to 1966, marks a period when scientific discoveries began to translate into widespread practical application. For example, quantum theory found tangible expression in industrial technologies like the laser. At the same time, questions about the risks of new technologies began to emerge alongside this progress. The nuclear weapons tests of the late 1950s and early 1960s demonstrated technology’s power but also its dangers.
This decade thus highlights an enduring challenge: how to translate scientific breakthroughs into economic value while identifying and mitigating unforeseen risks.
Lasers bring quantum technology to the mainstream
Before the 1960s, developments in quantum theory were largely seen as abstract and theoretical. That changed with the invention of the laser, which made it possible to produce highly focused beams of light with consistent intensity. This enabled industrial companies to cut materials with extreme precision, as well as unlocked new forms of medical treatment. Lasers are a key example of a scientific concept making the jump to industry. They are now used in countless applications—from fiber optics in the telecommunications sector, to barcodes, laser surgeries, and data storage.
For technology leaders, the quantum-to-laser pathway illustrates how quickly theoretical advances can become commercially relevant. However, recognizing this potential is challenging. It requires a clear understanding of the underlying technology, but also the ability to identify the use cases where it could solve real business problems affordably and at scale. Other applications of quantum theory, such as quantum computing, are still not widespread, showing that the leap from theory to application is not always a straight path.
Computing becomes smaller and more accessible
Another defining development of the decade was the introduction of the integrated circuit, which enabled the miniaturization of electronic components. Computers, which had previously occupied entire rooms, began to shrink in size. Systems such as IBM’s 1401, introduced in 1959, made computing more practical for business use and were widely adopted across industries. At the same time, early programming languages such as Fortran and COBOL made it easier to instruct machines to perform complex tasks. This combination of hardware and software advances allowed computing to move into the industrial and business applications we know today.
The implications were significant. Tasks that had required manual calculation could now be automated. Industries such as banking and accounting began to adopt computing to improve efficiency and accuracy. Over time, entirely new industries emerged around software and hardware development. However, many of these early systems were still purpose-built to solve specific problems. As a result, computers were often one-off machines that didn’t communicate effectively with one another, limiting their impact on business-wide value creation.
For today’s technology leaders, echoes of these early computers live on inside modern tech stacks. Many legacy systems in place today trace their origins to hardware and software decisions made during this period. The early focus on deploying computers for immediate use cases, without taking a broader architectural view, created much of the enterprise technology complexity that companies still face today. And tech leaders are still making these types of isolated decisions, such as deploying AI in a piecemeal approach instead of rethinking their architectures from the ground up for the agentic age. Every tech leader knows the challenge: They must build a modern technology infrastructure that reduces technical debt, while balancing short-term impact with long-term returns. The past shows just how hard that is to achieve.
The unintended consequences of chemical innovation
The decade from 1956 to 1966 also saw the widespread use of chemicals in agriculture and public health. Substances such as DDT, identified as an effective insecticide in the late 1930s and used during World War II to control malaria and typhus among troops, were deployed in this decade to increase crop yields and combat disease more broadly. At the time, these applications were widely seen as clear progress.
Over time, however, the side effects of DDT and other widely deployed chemicals could not be ignored. The persistence of chemicals in the environment and their impact on wildlife and ecosystems became apparent.
This period highlights an important lesson: Technologies that deliver immediate benefits can also introduce longer-term risks that are not visible at the outset. Understanding these risks requires ongoing assessment and a willingness to revisit earlier assumptions. In the age of AI, this lesson is especially relevant. In the breakneck race to deploy AI at scale for broad business benefit, it’s also critical that technology leaders consider its potential risks: security breaches, the spread of false information, and the human impact of job displacement, for example. Responsible AI strategies can allow companies to embrace opportunity while mitigating risks.
Progress with guardrails
The development and testing of nuclear weapons accelerated during this period, with large-scale testing programs conducted by major powers. The United States built on earlier tests such as the 1954 Bikini Atoll detonation by continuing high-yield atmospheric testing at the Nevada Test Site in the late 1950s and early 1960s. These tests had significant environmental and geopolitical consequences, with mounting concerns about radioactive fallout. The Partial Test Ban Treaty in 1963 marked an early effort to limit the use and spread of nuclear technologies.
This illustrates that not all technological capabilities should be pursued without constraint. In some cases, governance and regulation are necessary to ensure that the broader impact remains manageable. For today’s technology leaders, this lesson remains relevant. As agentic AI becomes more powerful and woven directly into business systems, enacting guidelines around appropriate use and human oversight of agent output are critical. Effective governance need not be a brake on innovation; it can instead be an accelerator for sustainable adoption.
A way forward: Responsible growth
The decade from 1956 to 1966 saw many scientific discoveries become part of everyday life. For many people in developed economies, technology was seen as a force for good. But toward the end of the decade, cracks in that optimism began to show—especially as the negative impacts of chemicals and nuclear tests became apparent.
Every technology leader knows that understanding how technologies are used, how they interact with existing systems, and what risks they introduce is just as important as the breakthroughs themselves. But in periods of intense technological disruption like the late 1950s—and today with AI—companies can overindex on experimentation without constraint. Yet, racing headlong into AI without first building an agentic organization will backfire. That’s why forward-thinking technology leaders are rewiring their infrastructures with AI at the center and reskilling their people to use AI as a true superpower.
The midcentury period in the United States and other developed economies was an era of profound growth and optimism that’s echoed in today’s excitement around AI. But the period also saw people grow disillusioned with breakneck progress that negatively impacted the planet and people. Today’s concerns about AI’s potential risks to society and the environment are thus a familiar refrain. For technology leaders, success with AI will depend not just on technology adoption, but on applying it in ways that are scalable, responsible, and aligned with long-term economic and societal objectives.
Chandrasekhar Panda is a partner in McKinsey’s Riyadh office, Henning Soller is a partner in the Frankfurt office, Klemens Hjartar is a senior partner in the Copenhagen office, and Sven Blumberg is a senior partner in the Istanbul office.