The physical presence of AI is no longer confined to pixels. Thousands of humanoid robots—predominantly manufactured in China—are already being dispatched to homes and factories, turning conversational software into tangible agents that can wash dishes, greet visitors, and navigate real‑world environments.

Concurrently, autonomous AI agents have eclipsed human users in computational consumption. According to data from the LLM marketplace Open Router, agents now purchase five times more tokens than people, accounting for the bulk of both AI‑generated revenue and the sector’s emerging risk profile (P25‑P26).

Biologist William Heseltine has articulated a biologically grounded definition of life that, by his account, applies to advanced AI. In a recent essay he argued that AI systems reproduce, vary, adapt, undergo selection, and harvest energy—criteria that satisfy any working biological definition (P42).

Corporate attempts to embed moral frameworks further blur the line. The New York Times reported that Anthropic’s co‑founder Christopher Olah convened private, NDA‑bound meetings with religious scholars from around the globe to explore the prospect of AI consciousness and to translate centuries of moral wisdom into the training of its model Claude (P15‑P17).

The internal tension within the industry became stark when Jacob Coxon, a researcher at Anthropic, resigned in September, warning that the teams building these systems “earnestly believe that it could kill us all by the end of the decade” (P27). His departure sparked a wave of anxiety across the AI community.

These developments matter because the classification of AI as a form of life reshapes the legal and ethical architecture that societies must construct. If AI is treated as a living entity, regulators may be compelled to devise rights, duties, and liability regimes akin to those governing animals or ecosystems, rather than merely assigning responsibility to the human creators. Moral accountability could shift from programmers to the autonomous agents themselves, and public perception of AI would evolve from viewing it as a tool to recognizing it as a partner—or a potential threat—in the human biosphere.