Earlier this month, former Anthropic employee Jacob Coxon shocked the public with a stark and unsettling warning: "The people building artificial intelligence earnestly believe that it could kill us all by the end of the decade."
This dire prediction thrusts the existential risks of advanced technology back into the global spotlight. Yet, beyond the immediate alarms raised by software engineers and industry executives, the breakneck rush to develop increasingly powerful artificial intelligence forces a deeper, more uncomfortable reckoning. It raises fundamental questions about how modern society understands human nature, our mutual responsibilities to one another, and the kind of world we are actively constructing.
To explore these profound intersections of technology, ethics, and human identity, Broadview spoke with Hyejung Yum, a professor at Concordia University in Montreal whose research focuses on theology, postcolonial and decolonizing theology, religious ethics, and technology.
From Software Engineering to Religious Ethics
When asked about her initial reaction to warnings from Anthropic researchers and executives that AI could potentially destroy humanity within the next decade, Yum drew upon her unique dual background. Before pursuing advanced studies in theology and religion, she worked as a software engineer, granting her an insider’s perspective on the mechanics of technological innovation.
Technology, including artificial intelligence, is already woven deeply into the fabric of daily life. For instance, global communication and academic discussions now routinely take place across digital video platforms rather than in physical rooms. However, contemporary AI introduces an entirely unprecedented set of questions that set it apart from previous human inventions.
Yum points to remarks made by Anthropic co-founder and research lead Christopher Olah when commenting on papal writings regarding AI. Olah admitted that even as a lead researcher, he sometimes could not fully understand the internal processes of the advanced systems his team helped create—a reality he found deeply unsettling.
Unlike historical technologies such as airplanes, which were engineered through the deliberate fabrication and assembly of individual, understandable parts, modern AI systems are built differently. Humans now create simulated environments where algorithms learn autonomously through vast datasets and recursive learning loops. This capacity for independent adaptation makes modern AI inherently unpredictable. According to Yum, that unpredictability poses a tangible threat to humanity unless robust regulations and governance frameworks are established to ensure that technological progress firmly serves the common good.
The Technocratic Paradigm and Human Worth
The rapid emergence of machines capable of outperforming humans intellectually challenges long-held assumptions about what makes human beings valuable. This dilemma has increasingly drawn the attention of religious and ethical leaders worldwide.
Pope Francis addressed this trend in his encyclical letter Laudato Si’, warning against what he terms the "technocratic paradigm"—a dominant worldview that prioritizes technological power, efficiency, optimization, and absolute control as the ultimate solutions to complex human and social problems.
Building on this critique, contemporary scholars note that when the technocratic paradigm shapes societal values, the fullness of life is often falsely equated with having more possessions, minimizing weaknesses, eliminating uncertainty, and exerting total control over every variable. In this utilitarian framework, human beings risk being reduced to mere projects to be optimized rather than recognized as persons inherently called to relationship, community, and mutual communion.
While artificial intelligence may have been initially conceived to assist human flourishing, the underlying assumptions driving its development can quickly curdle into the belief that human vulnerability is a defect to be systematically overcome. In stark contrast, disability theologian Thomas Reynolds views human vulnerability not as a limitation to be excised, but as an essential opportunity to connect with others and build meaningful community.
Under the sway of the technocratic paradigm, however, human limitations, emotional frailties, or cognitive differences are frequently treated as inferior traits or obstacles that need to be engineered away. Yum argues that this mindset becomes deeply dangerous if society continues to accelerate AI development without thoughtful public policies, transparent oversight, or accountable ethical systems. Because the current race for artificial general intelligence is heavily driven by geopolitical competition and the pursuit of corporate profit, it operates with remarkably few binding ethical guidelines. Ultimately, this dynamic forces a critical question: Through our relentless pursuit of AI, what kind of humanity are we actually cultivating?
Redefining Relationships: I-It vs. I-Thou
Beyond immediate concerns regarding safety, systemic bias, or economic disruption, the evolution of artificial intelligence forces a fundamental reexamination of how humans define their own worth. Modern society’s valuation of individuals remains stubbornly tied to narrow definitions of rationality, productivity, and raw cognitive intelligence—metrics that algorithms are increasingly optimized to surpass.
This narrow focus ties back directly to the technocratic worldview. To better understand the nature of human connection in a digital age, Yum points to the framework established by Jewish philosopher Martin Buber, who distinguished between "I-it" and "I-thou" relationships.
An "I-it" relationship involves viewing others merely as objects to be categorized, manipulated, controlled, or utilized for a specific function. Conversely, an "I-thou" relationship represents a profound encounter where individuals experience one another as whole, irreducible persons, rather than reducing them to mere instruments or functional utilities.
Currently, human interactions with artificial intelligence lean heavily into the "I-it" category, treating machines as tools designed to serve commands and maximize output. Yet, as the global development of AI becomes inextricably bound to a high-stakes geopolitical competition for technological dominance, the "I-it" relational model begins to bleed over into human-to-human interactions. When efficiency and control become the primary metrics of interaction, human value risks being measured by those exact same cold standards.
The core crisis, Yum suggests, is not solely located within the technology itself, but rather in how this dominant ideology shapes the way humans perceive themselves and one another. As religious leaders and ethicists have increasingly warned, unchecked allegiance to the technocratic paradigm can easily steer society toward an anti-human vision, where the essence of human dignity is eclipsed by the relentless demand for optimization and control.