AI · November 10, 2026
Anthropic Chief Dario Amodei's Long-Standing Focus on Artificial Intelligence Risks
Dario Amodei, the chief executive of the artificial intelligence company Anthropic, has maintained a deep concern regarding the potential dangers of the technology for nearly two decades. His conviction that artificial intelligence could eventually pose a threat to humanity was formed long before he established his current organization. Two distinct intellectual currents have shaped his perspective and provided the foundation for the company he founded in 2021.
A reconstruction of Amodei's background, based on interviews with more than twenty friends, classmates, teachers, and former colleagues, portrays him as a figure who views artificial intelligence not merely as a technological advancement but as an existential threat requiring careful management. His early exposure to these ideas occurred during his student years. In 2007, while at Princeton University, he discussed the book The Singularity Is Near with a friend. The book predicted that artificial intelligence would eventually surpass human intelligence, a possibility Amodei believed could occur within his own lifetime.
His intellectual framework was heavily influenced by online communities discussing artificial intelligence risks and methods to improve the world effectively. Two specific movements played a significant role in his thinking: rationalism and effective altruism. Rationalism focuses on the objective application of logical and mathematical reasoning, holding the belief that artificial intelligence could become powerful enough to destroy society. Effective altruism centers on the efficient use of resources to improve the well-being of as many people as possible. These ideas converged in Amodei's belief that only those who take the risks seriously should develop the technology, a conviction that reportedly contributed to his decision to enter the artificial intelligence industry.
After completing his doctoral research in physics, Amodei worked at Baidu and Google Brain, where he focused on artificial intelligence safety. In 2016, he moved to OpenAI, where he led the safety team and worked on developing increasingly powerful models. Tensions arose within OpenAI regarding the distribution of responsibilities for artificial intelligence safety. Amodei believed he was particularly suited to oversee the ethical aspects of the technology, while other employees expressed concern over the growing influence of him and his sister Daniela. In December 2020, the two left OpenAI along with several colleagues. One month later, they founded Anthropic, a company explicitly dedicated to the development of safe artificial intelligence.
Amodei's views have attracted both supporters and critics. Advocates within the artificial intelligence safety community believe he takes the technology's risks seriously. Critics, including leaders from Meta and Nvidia, accuse him of being overly pessimistic and seeking excessive control over the development of artificial intelligence. He has also voiced political opinions, advocating for restrictions on the sale of advanced chips to China to prevent the country from strengthening its position in the industry. Additionally, he supports regulations to prevent artificial intelligence systems from becoming uncontrollable. His stance has led to conflicts with the United States Department of Defense, as he has opposed the use of artificial intelligence for mass surveillance of Americans and the deployment of autonomous weapons. In an essay published in September, he argued for slowing the development of artificial intelligence, stating that the technology's benefits can only be realized if it is developed with extra caution.