For Kiara Nirghin, 24-year-old co-founder and chief technology officer of applied AI lab Chima, the narrative that her generation is using artificial intelligence as a cheat code is not only false: it ignores a fundamental shift in human cognition.
The former Stanford computer science student and Pierre Thiel A colleague argued that while older generations view AI as a tool worth adopting, Generation Z considers it a native language. However, with this mastery comes a unique burden: the “AI anxiety” of keeping pace with a technology that is currently the “worst” it can be.
Speaking at Fortune Brainstorm AI in San Francisco, Nirghin addressed the tension between Gen Z’s perception and their reality as builders. “The truth is that the younger generation is not adopting AI,” she said. “We grow by mastering AI.” This distinction is essential in the workplace. While a manager might see an employee using an AI agent to speed things up, Nirghin said she sees a change in the architecture of work itself.
“We don’t plan to code from scratch,” she explained. “We are considering coding with a side-by-side coding agent.” Far from being a generational shorthand, Generation Z is a pioneer, she argued.
“It fundamentally changes the way you write, the way you take tests, the way you apply for jobs or different applications, because it doesn’t come from the ground up,” Nirghin said of working side-by-side with an agent. “I think it really means that the wide range of use cases and applications that we’re seeing are really being driven by the younger generation.”
The myth of “laziness” versus deep thinking
One of the most common criticisms of the digital native generation is that reliance on large language models (LLMs) erodes critical thinking skills. Nirghin firmly rejects him. “I think the biggest misconception is that young people are using AI to not think,” she said, and that they are using it “as a shortcut.”
Instead, Nirghin said, smart users leverage these tools to offload cognitive work so they can explore complex topics with greater intensity. She said it’s not as simple as putting the “cognitive load” on an AI model, it’s about thinking “differently…even more ‘deeply’ about a specific topic, because the agent frees you from hours of menial work.”
As an example, she highlighted completing in-depth research reports on financial markets that could take hours to generate manually. By automating this work, she said the user is free to analyze the implications rather than just collect the data. “What does this unlock for you?” she asked the audience, inviting them to think about how much more they can do with these tools at hand.
The anxiety of infinite improvement
Nirghin said his generation faces a daunting reality that people don’t appreciate: the relentless speed of obsolescence and their own awareness of it. She said fears about AI have some similarities with “climate anxiety.” Noting that some of her early research focused on climate change, she explained climate anxiety as the idea that “a climate change movement is coming and we don’t really know what to do, but we know it’s coming and no one is moving that quickly to solve the problem.” »
This is linked to the realization that current technology, as impressive as it may seem, is primitive compared to what comes next. “The current models are as stupid as they will ever be,” Nirghin warned. “It will only get faster, more advanced and smarter, with every model from now on.”
For Gen Z workers, she said, this creates a pressure environment in which staying ahead of the curve is a daily requirement. Nirghin noted that recent model releases have “gobbled up SKUs in such a huge way” that previous capacities can now be “increased 10-fold” overnight – imagine coming to work tomorrow, capable of producing 10 times more than yesterday. If an employee isn’t consistently aware of these updates, “you’re kind of left behind.” The fear is not taking too many shortcuts, but not determining every path and every update to reach that factor of 10.
Taste like the new IQ
If intelligence is trivialized by models that improve exponentially, what becomes the new measure of human value? According to Nirghin, it is “taste”.
Nirghin, whose experience includes working at Stanford Human-Centered AI Labsargued that accuracy criteria no longer reflect what makes a product successful. She cited the example of coding agents who, without human guidance, might uncontrollably add “glitter emojis” to a front-end user interface because they “like” certain design tropes.
“You know something is vibe coded if you’ve ever worked with a coding agent,” she joked. The differentiator for tomorrow’s workforce will not be the ability to generate code or text, but the human-centered judgment to determine what users actually want to see. “As models, use cases and efficiencies change,” Nirghin said, “the key differentiator is taste.”
Nirghin’s advice extends beyond his peers and to the older generations currently leading them. She emphasized that “mastering AI is just as important for people who are already in the workforce,” urging them to arm themselves with tools like ChatGPT or Gemini as their daily “co-pilots.”
Ultimately, Nirghin said she sees the rapid evolution of AI not as a threat to jobs, but as a challenge to adapt. Whether it’s the automation of back-office processes or the launch of “deep search agents”, the economic “unlock” provided by these models is already incredible, even if they have never improved. But the anxiety of keeping up is the new price of admission to the future of work.
