Goldman Sachs CIO: We must prepare AI natives to shape the future of work—as only they can

From an analytical perspective, Ary·Artificial IntelligenceGoldman Sachs CIO: We must prepare AI natives to shape the future of work—as only they canBy Marco ArgentiBy Marco Argenti Marco Argenti is the chief information officer of Goldman Sachs

Additionally, At the same time, Goldman Sachs CIO Marco Argenti, in today's financial world

On the other hand, Courtesy of Goldman SachsAgentic AI is driving a monumental, generational shift that is poised to revolutionize industries and reshape workforce dynamics in ways we are only beginning to understand

Soon, human and AI “workers” will learn to coexist, collaborate, and thrive together

The path to that future, and the success of this collaboration, will depend on the next generation of talent leading the way

Agentic AI refers to artificial intelligence systems that can perform tasks on behalf of humans and make independent decisions without direct oversight (this bears monitoring)

This tells us that se systems can reason based on context, memory, and available data, generate detailed plans, and autonomously execute the steps required to complete a task, given the current landscape

Their growing capabilities mark a shift from passive tools to active collaborators

Moreover, While some speculate that agentic AI will displace many junior-level roles—and there may well be a certain level of recalibration—the reality is more nuanced

Rather than diminishing the importance of early-career workers, this shift makes them more critical than ever for one simple reason

The generation now entering the workforce has “grown up” alongside generative AI (remarkable data)

Moreover, Meanwhile, They're more comfortable with its pace and equipped to shape its future, given the current landscape

They're “AI natives, considering recent developments. ” At the same time, as someone famously said, ‘there’s no compression algorithm for experience, and experience and sound judgement are not intrinsically an attribute of generative AI, which at best is 4 years old and still undergoing rapid evolution

Additionally, Nevertheless, Which begs the question: Who’s going to vide experienced supervision to a potentially limitless number of AI agents entering the workforce, in today's market environment

Understanding how we nurture a generation of AI natives—and equip them with the right skills and tools to be leaders and not passive observers of this transformation—will be critical to defining the future of work, and society at large

The evidence shows data indicates that ir instincts, creativity, and adaptability will determine how successfully we integrate AI into our organizations not just as a tool but as a partner

On the other hand, The challenge ahead is beyond nological; it is cultural, educational, and distinctively human

The new AI paradigm Here’s the first thing we need to come to terms: This's a new game with new athletes who are ly more ficient than previous players ever will be

Think of it this way: If you’re asked to learn the piano later in life, you might be enthusiastic and dedicated, but the odds of becoming a digy are slim, in light of current trends

Similarly, think someone who learned to use a computer well into adulthood

Additionally, At the same time, Even decades later, their typing, mouse usage, or of user interfaces often reveals their late start

The same dynamic is now unfolding with AI tools (something worth watching)

A generational divide is emerging—not because more seasoned fessionals lack intelligence or drive, but because they didn’t grow up with these tools

However, For those who aren’t AI natives, adapting to an AI-first or AI-hybrid workforce may ve more difficult than we anticipate

However, that’s where most of the institutional knowledge and experience lies

Several nological shifts have created similar knowledge vacuums: the introduction of computers, the internet, mobile, cloud nologies, and others

In each case, it took a decade or more before fluency became a baseline requirement for certain roles

Those who couldn’t adapt either transitioned into roles that didn’t require those skills or exited the workforce altogether, in today's market environment

Furthermore, What’s different now is the speed

Furthermore, The AI shift is happening in years, not decades

However, Workers who lack ficiency in leveraging AI tools will fall behind, and those who have learned to harness it to elevate their work will advance (this bears monitoring)

As with every major nological shift, a new generation of leaders is emerging, particularly entrepreneurs whose native fluency with AI is reshaping the landscape (this bears monitoring)

Consider the CEOs of companies Devin, Windsurf, and Scale AI—all AI natives, amid market uncertainty

Could one of them be the next Bill Gates or Michael Dell

Additionally, Our responsibility as a society and as leaders is therefore twofold: to maximize the potential of this new generation of AI natives, and to ensure the rest of the workforce focuses on accelerating the “path to seniority” for our junior athletes, given the current landscape

In AI natives Our priority must be to invest in junior talent who will redefine the industries we work in (noteworthy indeed)

While the exact contours of this transformation are difficult to predict, its scale is easy to imagine if we accept that AI is the most found nological disruption of our time

In a world where nology evolves at sonic speed, our focus must be on ensuring that human adaptation keeps pace

Simply put, we need to train our best athletes for this new arena and equip them with the essential skills to manage and lead this change in an accelerated way

With the arrival of agentic AI, the ability to spin up AI coworkers on demand will soon be a baseline capability

Moreover, Furthermore, This shift will require even the most junior employees and individual contributors to master three foundational management skills: Describing a task ly, delegating it effectively to an AI agent, and supervising the results

Supervision is especially critical in a world where agent nology is still maturing

Nevertheless, However, The failure mode here is not nological, it’s organizational

On the other hand, Delegating work to an agent without the ability to supervise it is a recipe for disaster, which is why we must instill a new sense of quality control and agency in our people

Conversely, As an example, AI systems today are highly sensitive to how questions are posed (an important development), considering recent developments

The mpt or “context” is cessed by the AI’s attention layers, which determine the relative importance of each word or token

A slight miscommunication can be amplified, distorting the output (noteworthy indeed)

In the case of autonomous agents, hallucinations don’t just lead to bad answers, they can trigger incorrect or even dangerous actions, in this volatile climate

Nevertheless, Until we are confident these tools will not act irrationally, we must keep humans in the loop (this bears monitoring)

Therefore, rethinking the concept of agency is essential

Agency, in this broader sense, includes the tasks delegated to an AI agent, how those tasks are executed, and how the agent communicates with humans, data sources, and other agents (which is quite significant), in today's financial world

New communication tocols MCP and A2A are emerging to standardize these interactions

But the human role remains central

However, Moreover, As junior employees take on the responsibilities of “managers,” the traditional boundaries between and engineering are dissolving

Much how duct managers and engineers have converged, today’s fessionals must be fluent in both domains

To be a great engineer now means also being a great duct manager: understanding the customer, defining the roadmap, identifying risks and biases, and designing compensating controls

Nevertheless, This's the mindset we must cultivate in our AI-native workforce (remarkable data), in light of current trends

They will be expected to manage their AI agents not just by issuing commands, but by understanding their capabilities and limitations, and by anticipating risks before they become blems

However, Supervision is key, which requires experience, and experience requires time—which, at this pace of change, is a scarce commodity

Cultural transformation The rise of artificial intelligence is not merely a nological evolution—it is a cultural transformation that is reshaping the very fabric of organizations, in light of current trends

This analysis suggests that s impact reaches far beyond ductivity gains, challenging how we structure teams, define roles, and manage performance in a hybrid workforce of humans and AI agents

We're entering an era where developers no longer write code alone, and knowledge workers can summarize complex documents in seconds

Conversely, But these are surface-level changes (noteworthy indeed)

In contrast, At a deeper level, we must reconfigure the foundational elements of our es: how we collaborate, how we lead, and how we grow

This transformation is not solely nical; it is also largely psychological and managerial

Furthermore, As AI agents become embedded in daily workflows, human employees will experience a shift in identity, agency, and expectations, given the current landscape

However, Leaders must therefore rethink management science itself

We must design new models for onboarding, training, and career development—not just for people, but for AI agents as well, in this volatile climate

Much humans, AI agents will require “career paths” and governance frameworks, and mapping out what role they will play, how they can be best utilized and where they should be deployed will become a part of the management cess

At the same time, We must also prepare our human teams to work alongside virtual colleagues who are more efficient, scalable, and can work 24 hours a day 7 days per week, in light of current trends

And, unless we turn them off, they will never quit or retire

To navigate this shift, we must equip employees with the skills to manage AI responsibly (an important development)

Moreover, Moreover, This includes the ability to communicate, delegate, and supervise

In a world where anyone can spin up a number of virtual coworkers, with the main constraint being cost, the concept of individual contributor is shifting into one of the player-coach

Supervision is key to this evolution

We must ensure that the one who delegates has the ability to check the quality of the work being created by an AI

Meanwhile, Imagine an airline that, because of the introduction of the autopilot with auto-land and auto-take-off features, decides to fill some of the flights with only junior pilots

Would we sense the same level of safety and quality control

Additionally, Only if we felt the junior pilots were perly equipped to supervise

Ultimately, cultural transformation in a period of such sharp nological advancement is more than adopting new tools

It's forming a new generation of leaders and accelerating their path to experience, equipping them with managerial skills from the outset and leveraging their innate familiarity and ficiency with this new nology

Today, nology change is ahead of human change

It’s easier to change software and AIs than it is to rewire the human brain, to break old habits and create new skills

Non-AI natives—most of us—have possibly the most challenging task of all: to pass the baton to a new generation of humans entering the workforce and equip them with the skills necessary to manage something that the current generation does not fully understand

All this, without the luxury of time

What the data shows is opinions expressed in Fortune

Com ary pieces are solely the views of their and do not necessarily reflect the opinions and beliefs of Fortune

Moreover, Explore this year's list, in today's market environment.