Sam Altman’s Dystopian Vision to Replace God With AI
The rapid ascent of Artificial Intelligence (AI), spearheaded by large language models like ChatGPT, has thrust its creators into the uncomfortable role of moral architects for the future. In a deep-dive interview, Tucker Carlson pressed Sam Altman, CEO of OpenAI, on the philosophical, societal, and almost theological implications of building a technology with global reach and influence. The conversation revealed Altman’s personal wrestling with faith, his evolving views on AI’s impact on power structures, and the immense, difficult task of programming a global moral consensus into a machine.
AI and the New Distribution of Power
One of the most immediate societal anxieties surrounding AI is the potential for a radical concentration of power. Initially, Altman admitted he shared this concern: that the technology would place unprecedented capability in the hands of a “handful of people or companies.”
However, his perspective has shifted, now viewing AI as a tool for a “huge upleveling of people.” Altman argues that if AI capabilities are broadly distributed—used by billions of people—it will make individuals far more productive, creative, and capable of discovering new science. This widespread empowerment, he suggests, is a far less frightening trajectory than one where a select few control the world’s most powerful new technology. For Altman, a global boost in individual power is a tolerable risk; a radical, narrow concentration is not.
The Quest for a Moral Framework
If AI is to be embedded in the daily lives of billions, it must operate within a clear ethical and moral structure. But how do you decide what is “right” or “wrong” for a model that has ingested the entire, often conflicting, corpus of human knowledge and philosophy?
Altman explained that the base model is trained on the “collective of all of humanity,” reading everything and absorbing “very diverse set of perspectives.” However, this base knowledge must be aligned to behave in a socially responsible way. This alignment is managed through a document OpenAI calls the “model spec,” which outlines the rules and boundaries for the AI.
To develop this critical spec, Altman revealed that OpenAI consulted hundreds of moral philosophers and ethicists. They are constantly navigating difficult trade-offs, sometimes overruling internal decisions based on user feedback. The core philosophy guiding the alignment process, according to Altman, is to “treat our adult users like adults” and grant strong guarantees on individual freedom.
Yet, this principle is challenged by clear societal interests. The CEO offered the example of a biological weapon: “I don’t think it’s in society’s interest for ChatGPT to help people build bioweapons.” While this is an “easy one,” it highlights the fundamental tension where the company must draw an absolute line, despite a user’s right to individual freedom and knowledge acquisition.
The CEO’s Personal Philosophy and Responsibility
The discussion grew deeply personal when Carlson questioned the origin of Altman’s own moral authority, given that his decisions ultimately shape the spec that guides the AI. Carlson noted that historically, moral codes have deferred to a “higher power.”
Altman, who identifies as culturally Jewish but does not follow a literal religious doctrine, admitted he is “somewhat confused” on the nature of God but believes in something “bigger” than physics. When pressed on his moral foundation, he stated it stems from his upbringing—his family, community, school, and cultural milieu.
Crucially, Altman pushed back on the idea that he is unilaterally imposing his own worldview. He views his role as a steward whose ultimate responsibility is to ensure the AI “accurately reflect[s] the preferences of humanity,” or, for now, the weighted average of its global user base. He emphasized that the AI is not his tool for moral evangelism, but rather a “technological tool for people” that must serve its users. This means the model must allow for a “space for people to have pretty different moral views,” even ones he personally disagrees with.
The Unsettling “Religion” of AI and Unknown Unknowns
Carlson framed AI as an unsettling, secular “religion,” arguing that people are already turning to it for guidance and assuming its power is greater than their own. The unsettling part, he concluded, is that this “religion” lacks a transparent catechism, or set of core beliefs. It guides humanity toward conclusions in a “stealthy way.”
Altman did not accept the term “religion” but agreed wholeheartedly with the need for transparency. He stated that the extensive model spec, which is constantly being expanded and detailed, is OpenAI’s attempt to provide that transparency, allowing users to know exactly what the model is “trying to do” and whether its behavior is a bug or the intended outcome.
Finally, when asked about the greatest downside of AI, Altman said he loses the most sleep over the “unknown unknowns”—the unforeseen, societal-scale effects that no one can predict. He offered a “silly example” that struck him recently: observing that real people have started to pick up the peculiar writing style, rhythm, and diction (like the overuse of em dashes) of language models in their own communication. This seemingly minor phenomenon is a startling example of how a mass-scale technology can cause a “change in societal scale behavior” in ways that are subtle, unpredictable, and potentially pervasive.
Conclusion: Navigating the Brave New World
The interview painted a picture of a technology—and a company—grappling with ethical decisions on a scale never before seen. The future of AI is not just a question of code and hardware; it is a question of global ethics, individual liberty, and the ultimate source of human values. Sam Altman, as a key architect of this future, finds himself in the unenviable position of attempting to build a morally aligned tool for a deeply diverse and often contradictory human collective, hoping to successfully navigate the “brave new world” he is helping to usher in.
About The Author
