Why human skills – but not the number of humans (sorry) – matter more as AI spreads at work

Psychology Today’s Laura Berger argues that as AI becomes more prevalent in workplaces, the solution to avoiding the “Uncanny Valley”—where AI-generated content feels eerily human but emotionally vacant—lies in strengthening distinctly human capabilities rather than making AI more human-like. The piece emphasizes that longstanding relationships, metacognition, emotional intelligence, and adaptive momentum are irreplaceable human assets that become more valuable, not less, in an AI-driven world.

The big picture: The concept of the Uncanny Valley, originally applied to human-like robots that provoke discomfort, now extends to AI-generated workplace communications that hit technical marks but lack emotional resonance.

• Berger suggests that instead of trying to make AI more human, we should focus on making our human systems “more connected, self-aware, and emotionally intelligent.”
• The article positions AI as most powerful when used alongside uniquely human capabilities rather than as a replacement for them.

Key human advantages: Four distinct areas where humans maintain irreplaceable value in professional settings emerge from the research.

• Historical insight: Long-term relationships provide years of context that help recognize genuine change and identify potential pitfalls.
• Emotionally attuned support: Familiarity enables nuanced understanding of when to push forward and when to pause.
• Rich synthesis: Trusted colleagues hold “decades of data AI doesn’t” and can see connections that transcend algorithmic training.

Why metacognition matters: The ability to reflect on thought processes—something AI lacks—becomes a leadership imperative as AI embeds deeper into decision-making.

• Metacognition involves stepping back to ask questions like “Why am I approaching it this way?” and “Is this belief still serving me?”
• AI doesn’t know when its predictions are wrong or whether its patterns have unintended consequences, making human oversight essential.
• Good metacognition often emerges from dialogue with trusted colleagues who can mirror not only our ideas but our mental habits.

Why emotion matters: Cognitive neuroscience reveals that separating emotion from logic creates a false divide, as emotion actually drives cognition and fuels attention, motivation, memory, and behavior.

• AI tools can mimic empathy in tone but lack true emotional processing, creating gaps in feedback delivery and relationship building.
• “Emotional data often lives beneath the surface: a pause before a sentence, a shift in tone, a glance that says more than words.”
• Emotionally intelligent conversations that change minds or hearts are built through trusted relationships over time.

Building trust through movement: Trust emerges not from perfection but from consistent, emotionally engaged interactions that demonstrate adaptability.

• The neuroscience of trust suggests that collaborative iteration—a distinctly human strength—reinforces trust more than the illusion of certainty.
• “We can reflect, revise, and respond. We can change our minds and say so. That kind of progress reinforces trust more than the illusion of certainty.”

What leaders should focus on: Berger recommends four specific areas for AI preparation that center on human capabilities.

• Strengthen existing relationships and identify underutilized sources of strategy and reflection.
• Sharpen metacognitive abilities to remain conscious of thinking processes.
• Attune to emotional signals that AI cannot detect or process.
• Prioritize adaptive momentum that creates shared understanding and transparency.

The bottom line: Rather than competing with AI on technical capabilities, the path forward involves deepening human connections and emotional intelligence to complement AI’s strengths while addressing its fundamental limitations.