Anthropomorphizing AI: The Next Frontier in Our 18,000-Year Journey with Technology

2. Cognitive Dexterity (3,000 years ago): Involves developing cognitive skills and understanding cause and effect, using tools like the bow and arrow. Humans devised tools that extended our physical abilities. Compared to atlatl hunting, Bow hunting with poisoned arrows, for example, symbolized a leap in thinking, a form of remote problem-solving that magnified our cognitive, conceptual, behavioral, and technological flexibility. Studies have shown statistically higher levels of neural activity in arrow shooting compared to spear throwing.^[2] The act itself required increased visual acuity, context updating, and memory load. But beyond technology, the adoption of bow hunting initiated profound social change. ^[3]Atlatl hunting was once a group activity. With bow and arrow, one could hunt alone, causing tectonic shifts in social norms and structures from who you married to your status within the group. In our current era, we face a similar challenge. We must master new digital tools that extend our capabilities, allowing us to solve problems remotely and visualize potential outcomes.

3. Aesthetic Dexterity (500 Years Ago): Reflects the expression of identity through tools and symbols, understanding human psychology and social dynamics. In the Iron Age, human creativity took a new form, tools began to carry symbolic meaning, communicating anthropomorphic features like gender, age, and status. Designing the outside appearance of tools and imprinting them with motifs became a way to communicate anthropomorphic features like gender, age, and status. These designs often mirrored human transformation ceremonies, reflecting societal values and individual identity. This was a profound way of expressing self and community, much like how we personalize and brand our digital interfaces today.^[4] Humans still see tools from an aesthetic perspective until today. Russell Neuman, a professor of media technology at New York University, points out that people often view AI as an autonomous entity— a robot or a self-serving machine—that poses a threat because it might soon outsmart us.^[5] This perception arises because we tend to anthropomorphize AI, projecting human traits onto these evolving technologies.^[6] For example, the conversational nature of ChatGPT provides it with a persuasive social presence, making it difficult not to imagine there’s a mind behind the screen. As a highly social species, we are inclined to attribute human-like qualities to various entities. This inclination to assume intentionality doesn’t start or end with artificial intelligence. Consider how a toddler might talk to their favorite toy truck, believing it can think, feel, and respond. The child might imagine the truck being happy when it’s being played with or sad when it’s left alone. This behavior demonstrates the toddler’s natural tendency to attribute human-like emotions and intentions to inanimate objects.

4. Machine Dexterity (60 Years Ago): Emerges in the Electronic Age, emphasizing a conceptual understanding of machines’ internal mechanisms, like silicons and transistors. As we entered the Electronic Age, technology became an integral part of everyday life. Devices like light bulbs, telephones, radios, and TVs were no longer mere conveniences; they were expressions of modern living. This was not just about aesthetics as the Iron age; it was about making the invisible visible, connecting the inner workings of machines with human experience. Humans wanted to know how these silicon transistors look from the inside. As this was something that demanded an explanation, a visual interpretation that would connect with both business audiences and the public.
This challenge led to a collaboration between art and technology. Artists were commissioned by big tech companies like IBM to create works that explained and illustrated these scientific and technical concepts. In her book,  Megan Prelinger stated that “Artists bridged the gap between invention and understanding, between business and industry, and between technology and the public”.^[7]

Today, we are faced with similar challenges as we implement AI, autonomous agents, and other digital tools in business. How do we explain these complex technologies? How do we make them accessible and meaningful to our citizens? The lesson from the Electronic Age is clear: We must engage with art, design, and human-centered communication to bridge the gap between technology and people.
Our journey through human history – from the Stone Age’s reasoning to the Iron Age’s symbolic expression, and the Electronic Age’s fusion of art and technology – teaches us valuable lessons for today’s leadership. Artists are crucial today in helping us understand AI’s inner workings, much like how people historically sought to understand the mechanics during the era of Machine Dexterity. For instance, artists like Mario Klingemann expose neural glitches in AI, revealing distortions and biases in algorithmic reasoning.^[8] This artistic mediation helps make AI systems more transparent and comprehensible, promoting responsible use and fostering a deeper understanding of the technology, echoing our historical curiosity and need to demystify complex machines.

Embracing our heritage of innovation, adaptability, and human-centered design will guide us in navigating uncertain times, building resilience, and fostering an organization that is in tune with the human experience.

5. Digital Dexterity (20 years Ago): Encompasses the manipulation of information and systems, requiring a holistic understanding of technology, humans, and information interaction. The Digital Age marks a profound transformation in our relationship with machines. As technology permeates our daily lives, the physicality of machines has become abstracted into data and systems. Artists and designers have redefined ‘the machine,’ focusing on its processes and products rather than its physical form.

In this era, information is not just a product of machines; it’s a tangible asset that can be manipulated, visualized, and understood. The role of the designer has evolved, shifting into the domains of user interfaces, user experience, and data visualization. The depiction of technology has moved from the physical devices to abstract ‘bits’ of data, a leap in our cognitive dexterity.


As we embrace AI, autonomous agents, and digital transformation, we must draw lessons from the Digital Age. Our task is to make abstract concepts accessible, to humanize technology, to foster collaboration across disciplines, and to cultivate a digital dexterity that balances technological readiness with ethical considerations and human values.”

6. Adaptive Digital Dexterity (Today and Future): The newest stage, characterized by two-way communication, dynamic interactions, real-time adaptation, conversational AI, sensory inputs, multimodal communication, and customized responses, similar to ChatGPT-4o’s capabilities. However, the future will not be limited to these capabilities; we will also see the rise of social learning and AI robots’ cultural evolution. Future household robots and those on construction sites should be able to adapt to human actions in real-time.

A few years ago, MIT researchers developed a method for non-coders to teach robots tasks through a single demonstration, allowing robots to learn from ambiguous human actions and subsequently teach other robots. ^[9]This human-like social learning skill might become the cornerstone for AI systems that can cooperate in our daily lives. Robots can’t deviate from the steps they’ve learned, and the entire learning process remains heavily dependent on human involvement. Recently, researchers published in Nature^[10] developed AI agents capable of imitating human behavior in real-time without a lot of pre-collected human data, using a method called few-shot imitation. This research demonstrated that cultural transmission could act as a bridge to adaptation, allowing agents to exceed the original demonstrator’s performance. By enabling AI to refine and adapt learned tasks autonomously, this new approach addresses the rigidity and dependency issues found in the earlier MIT study, paving the way for more flexible and independent AI systems.