Ronan Farrow has written substantially about AI, particularly the question of who controls increasingly powerful AI systems and whether those people can be trusted with that power.

His most important work is a major 2026 New Yorker investigation, written with Andrew Marantz:

“Sam Altman May Control Our Future—Can He Be Trusted?” 

What Farrow is arguing

The article is not primarily a technical examination of whether AI will literally destroy humanity. Instead, Farrow investigates the people and institutions developing the technology, especially OpenAI and its CEO Sam Altman.

The major issues he raises include:

• AI could become extraordinarily powerful. Farrow takes seriously OpenAI's original premise that AI might become one of the most powerful—and potentially dangerous—inventions humans have created. The New Yorker+1

• The people controlling AI therefore matter enormously. His investigation asks whether Altman and OpenAI's leadership can responsibly exercise that power.

• Safety versus commercial expansion: OpenAI was founded with an unusually strong emphasis on AI safety, but Farrow examines whether commercial pressures and the enormous investment required to build AI have changed those priorities. The New Yorker

• Government and military applications: Farrow and Marantz examine OpenAI's increasingly important relationships with the U.S. government and military. Their reporting discusses the controversy surrounding AI being used for autonomous weapons and mass-surveillance applications. The New Yorker

• Concentration of power: A central concern is that a relatively small number of technology executives and companies may acquire extraordinary influence over a technology with consequences for society as a whole.

• Trust and governance: Farrow investigated internal OpenAI documents, interviewed more than 100 people, and looked closely at the 2023 board crisis surrounding Altman. LinkedIn

One particularly striking aspect of the reporting is the comparison with J. Robert Oppenheimer. Farrow and Marantz point out that Altman himself and early OpenAI discussions repeatedly invoked the Manhattan Project as an analogy. The comparison is therefore both about technological power and about the responsibility of the people who develop it. The New Yorker

An important distinction

Farrow's position is more nuanced than “AI will destroy mankind.” His reporting is essentially asking a prior question:

If AI becomes extraordinarily powerful, what happens if the people controlling it are not sufficiently trustworthy, accountable, or constrained?

That is a somewhat different—and arguably more immediate—AI danger than the science-fiction scenario of a machine independently deciding to exterminate humanity.

If you're interested because of your earlier question about how AI could actually destroy mankind within a decade, Farrow's reporting is particularly relevant to the human-control pathway: AI becoming powerful while decision-making becomes concentrated in a handful of corporations, executives, governments, and military organizations.

---

If you want, I can:

• Ask for a summary of Ronan Farrow's key AI concerns

• Request comparison of Farrow's view with other AI thinkers

• Inquire about implications of Farrow's investigation for AI governance