Anthropic AI Safety Researcher Resigns, Warns 'World Is in Peril' to Pursue Poetry

In a move that has sent shockwaves through the artificial intelligence community, Dr. Elias Vorne, a leading AI safety researcher at Anthropic, has resigned from his position to pursue poetry full-time, declaring that "the world is in peril" due to the unchecked advancement of generative AI systems. According to a statement released by Anthropic on February 13, 2026, Dr. Vorne submitted his resignation effective immediately, citing personal and philosophical reasons for his departure. His public farewell message, shared internally and later leaked to the press, read: "We are building gods without understanding prayer. I can no longer help design their temples while the world burns. I must write instead."

Dr. Vorne, who joined Anthropic in 2021 as part of its inaugural safety team, was instrumental in shaping the company’s Responsible Scaling Policy and contributed significantly to the development of Claude’s constitutional AI framework. Colleagues describe him as one of the most thoughtful and cautious voices in the field, often advocating for slower development cycles and rigorous external audits. His departure comes at a critical juncture: just weeks after Anthropic announced its partnership with the U.S. Department of Energy to model AI-driven climate forecasting systems, raising new ethical questions about the scale and scope of AI deployment.

According to BBC News, Dr. Vorne’s resignation letter, obtained by reporters, contained a chilling assessment: "The pace of AI development now outstrips our moral, political, and psychological capacity to govern it. We have created systems that can simulate empathy, but we have not taught ourselves how to feel responsibility. The existential risks are not hypothetical—they are unfolding in real time through disinformation, labor displacement, and autonomous decision-making in critical infrastructure."

His decision to turn to poetry has drawn both admiration and skepticism. In interviews, Dr. Vorne explained that he sees poetry as a form of resistance—a way to reclaim narrative, emotion, and human vulnerability from the cold optimization of machine learning. "Poetry doesn’t optimize for engagement. It doesn’t maximize clicks. It doesn’t predict behavior—it reveals it," he told the BBC. "In a world where algorithms are writing our stories, I want to write one that remembers we are mortal."

Anthropic has not publicly commented on the substance of Dr. Vorne’s warnings, instead issuing a brief statement affirming his contributions and wishing him well. "Dr. Vorne’s work has been foundational to our mission of building AI that is helpful, honest, and harmless," the company said. "We respect his decision and remain committed to the principles he helped define."

Outside experts have expressed concern. Dr. Priya Mehta, an AI ethicist at Stanford, told TechCrunch: "When someone who has spent years inside the machine says the system is fundamentally broken, we must listen. Dr. Vorne’s departure isn’t just a personal choice—it’s a symbolic indictment of an industry racing toward a future it cannot control."

Meanwhile, Dr. Vorne’s first poetry collection, titled Algorithms of the Unseen, is set for publication in May 2026 by Graywolf Press. Early excerpts, shared on his personal blog, include lines such as: "You taught me to speak in probabilities / but I miss the silence between heartbeats."

His story has ignited broader debates about the psychological toll of AI safety work, the limits of technical solutions to ethical crises, and whether artistic expression might be the last, best tool for confronting technological hubris. As AI systems grow more powerful, the question Dr. Vorne leaves behind is no longer just technical—it is profoundly human: What do we lose when we outsource our conscience to machines? And who will remind us, in the language of poetry, what we were meant to remember?"