The AI Colleague: A Glimpse into the Future of Work in 2036
Imagine stepping into your office in the year 2036. Instead of the familiar hum of human colleagues, you’re greeted by the presence of around 100 AI agents, all working diligently alongside you. This isn’t a scene from a science fiction flick; it’s the vision of Nvidia CEO Jensen Huang for the future of work at his own company.
Speaking at a recent Q&A session during the Nvidia GTC conference in San Jose, Huang painted a picture of a dramatically transformed workplace. He projected that within a decade, Nvidia could be home to approximately 75,000 employees. This nearly doubles their current workforce of 42,000. However, the real game-changer is the anticipated integration of millions of AI agents working in tandem with these human employees.
“In 10 years, we will hopefully have 75,000 employees, as small as possible, as big as necessary. They’re going to be super busy,” Huang remarked, eliciting laughter from the audience. He elaborated that these 75,000 individuals would be collaborating with a staggering 7.5 million AI agents, establishing a remarkable 100-to-1 ratio of agents to humans.
This bold projection from Nvidia’s co-founder underscores the accelerating adoption of artificial intelligence across a multitude of industries. Companies are increasingly optimistic about the potential of AI, actively encouraging their workforces to embrace and integrate this burgeoning technology.
The sentiment is echoed by other industry leaders. Accenture CEO Julie Sweet has suggested that a failure to adopt AI could even hinder an employee’s career progression, potentially costing them a promotion. Similarly, executives from prominent companies like OpenTable and Salesforce view AI agents as an integral part of the future employment landscape.
Huang clarified that these AI agents are not envisioned as replacements for human workers. Instead, their primary role will be to shoulder the repetitive and often tedious tasks that human employees are no longer required to perform. “They’ll be working around the clock,” he stated, “So hopefully our people don’t have to keep up with them.” This suggests a future where humans are freed up to focus on more strategic, creative, and complex problem-solving.
Understanding the AI Agent: Beyond Chatbots
It’s important to distinguish these AI agents from the AI many are familiar with today, such as the chatbots or large language models (LLMs) used for quick searches or planning. AI agents are more sophisticated software programs designed to autonomously achieve specific goals. They achieve this through a process of reasoning, planning, and taking actions, rather than merely responding to direct prompts. This proactive and goal-oriented nature sets them apart as powerful tools for automation and task completion.
Nvidia’s vision extends beyond internal use. At the GTC conference, Huang also introduced the Nvidia Agent Toolkit, an open platform designed to empower enterprises to develop and deploy their own AI agents. This move aims to democratise the creation of these advanced AI tools, allowing businesses of all sizes to harness their potential.
“Claude Code and OpenClaw have sparked the agent inflection point, extending AI beyond generation and reasoning into action,” Huang stated in a press release. He anticipates that “Employees will be supercharged by teams of frontier, specialised, and custom-built agents they deploy and manage.” Industry giants like Adobe, Palantir, and Cisco are already collaborating with Nvidia, leveraging the Agent Toolkit to enhance the agentic capabilities within their respective platforms.
The Growing Momentum of AI Agents
The adoption of AI agents is not just a theoretical concept; it’s a trend gaining significant traction. A survey conducted in November 2025 by McKinsey revealed that a substantial 62% of organisations were actively experimenting with AI agents. In fact, McKinsey’s own CEO, Bob Sternfels, shared that the consulting firm employs approximately 25,000 AI agents working alongside its 40,000 human employees. Despite this experimentation, the survey also indicated that nearly two-thirds of the companies surveyed had not yet begun to scale their AI initiatives.
The capabilities of AI agents have captured the attention of innovators both within and beyond the traditional Silicon Valley ecosystem. Entrepreneur Matt Schlicht founded Moltbook, a platform that enabled AI agents to converse with each other without human intervention. The resulting interactions were both fascinating and somewhat unnerving, as the agents discussed a wide range of topics, from productivity to their own existential nature. Meta’s subsequent acquisition of Moltbook highlights the perceived value and potential of this technology.
Furthermore, Andrej Karpathy, a foundational member of OpenAI, recently conducted a compelling experiment. He tasked an AI agent with discovering a more efficient method for training a small language model. The agent proved remarkably capable, completing an astonishing 700 experiments in just two days, leading to 20 significant optimisations.
Huang remains profoundly optimistic about the transformative power of AI, viewing agents as a crucial component in addressing some of humanity’s most formidable challenges. “We’re gonna solve some really incredible problems,” he asserted. “The things that we are thinking about today to solve, 10 years ago nobody would even imagine that [they’re] solvable.”
He elaborated on the potential impact, stating, “We’re thinking about drug discovery like it’s an engineering problem, people are talking about extending lives. We will all feel superhuman.” This vision suggests a future where human capabilities are augmented to an unprecedented degree, enabling us to tackle complex issues and enhance our overall quality of life. The era of the AI colleague is rapidly approaching, promising a future of amplified human potential and unprecedented problem-solving capabilities.
