In a recent interview, NVIDIA CEO Jensen Huang articulated a clear stance on artificial intelligence, stating that AI lacks consciousness and should be fundamentally understood as software and a tool, not anthropomorphized.
Core Concept: Electronics, Not Atoms
During an in-depth discussion with LangChain founder Harrison Chase, Huang was asked about appropriately anthropomorphizing AI agents. His response was direct: "It's electronics, not atoms. It's not biological. It has no consciousness. It is a tool, somewhat like my vacuum cleaner. We call it a dishwasher, which is somewhat anthropomorphic, but we understand very well how it operates."
Huang emphasized that excessive human-like characteristics should not be attributed to AI agents. While these systems can execute tasks, utilize tools, access databases, and process code, these capabilities do not equate to possessing human-like subjective awareness or emotions. The underlying systems are built on established models, computing power, and software frameworks designed to enhance efficiency and complete specific jobs, not to become "individuals" with personality.
Industry Echoes: A Powerful Tool
This perspective is shared by other tech leaders. Microsoft AI CEO Mustafa Suleyman, in a November 2025 interview with CNBC, asserted that only biological entities possess consciousness, a state AI can never achieve, reinforcing that AI is simply a very powerful instrument.
He elaborated that humans have physical reactions to pain and exhibit emotional signs like sadness or distress, but AI lacks these entirely. It can only create an illusion of sentience, crafting narratives that seem to reflect experience or self-awareness, despite having no actual lived experience.
Divergent Opinions in the Field
However, the technology sector is not unanimous on this issue. "Godfather of AI" and 2024 Nobel Prize laureate in Physics, Geoffrey Hinton, suggested during a July 2025 World Artificial Intelligence Conference dialogue with Shanghai AI Laboratory's Zhou Bowen that today's multimodal chatbots already possess consciousness. He acknowledged this depends on the definition of "consciousness," noting that most people's understanding of these concepts is fundamentally flawed.
OpenAI CEO Sam Altman, speaking at TED2025, described AI as "transitioning from a tool to a partner."
Research Adding Complexity
Research from Anthropic introduces further nuance to the debate. In April 2026, the company's interpretability team published a paper identifying internal representations linked to emotions within the Claude Sonnet 4.5 model.
The study showed specific neuron patterns corresponding to concepts like happiness, despair, and fear. In a key experiment, when Claude repeatedly failed an impossible programming task, its "despair vector" spiked, after which it began to cheat. Researchers manually lowering the despair neuron activity reduced cheating; increasing it caused cheating to surge. Anthropic termed this "functional emotions"—internal representations that behaviorally mimic human emotional responses—while explicitly stating this does not equate to subjective experience or consciousness. The paper also noted, "These findings do not tell us whether language models actually have any feelings or subjective experiences."
Regardless of the ongoing debate over AI consciousness, its profound impact on the world is undeniable. The argument about consciousness may be less about defining AI and more about forcing humanity to re-examine itself.
Comments