What the Doomsday Clock actually measures

On 27 January 2026 the Bulletin of the Atomic Scientists moved the Clock to 85 seconds to midnight. It is the closest setting in the Clock's history, five seconds nearer than the 89 seconds announced in 2025 and the 90 seconds that held from 2023 to 2024.

In its statement, the Bulletin's Science and Security Board attributed the move to a combination of factors it described under the heading of a failure of leadership: the expiration of the New START treaty and the fraying of nuclear arms control, record-breaking climate trends, developments in artificial intelligence, and biosecurity concerns. Those are the Bulletin's stated reasons, and they belong to the Bulletin — we report the reasoning rather than adopting it.

The older settings still circulate as though they were current, usually attached to an article that was accurate when it was written. If you see 90 or 89 seconds presented as today's figure, you are reading a historical setting.

The Bulletin of the Atomic Scientists was founded in 1945 by scientists who had worked on the Manhattan Project. The clock first appeared on the cover of the Bulletin in 1947 as a graphic device — a way to communicate risk to people who were not going to read a technical paper about it.

The setting is decided by the Bulletin's Science and Security Board, in consultation with its Board of Sponsors, which has long included Nobel laureates. It is announced once a year rather than continuously, and the reasoning is published alongside it. Reading that statement is the entire point; the number by itself is a headline, not information.

The clock started as a nuclear indicator and has widened over the decades. The Bulletin's stated scope now includes nuclear risk, climate change, biological threats, and disruptive technologies including artificial intelligence — the categories of danger that are made by people and could, in principle, be unmade by them.

The seconds are rhetorical. They are not 85 literal seconds, 85 months, 85 years, or a probability rendered as time. There is no model that converts the Bulletin's deliberations into a duration, and nothing ticks between announcements: the hand sits where the board left it until the board meets again.

It is also not a forecast. It has moved away from midnight as well as toward it — most dramatically to 17 minutes in 1991 — which is the clearest available evidence that it is an assessment rather than a countdown. A countdown that can go backwards is not a countdown.

It says nothing, either, about asteroids, supernovae, or any other hazard that is not of human origin. Those are somebody else's department, and that department has telescopes.

The Clock is a single number attached to the largest possible subject, released on a scheduled date, with an image built for sharing. That is close to a perfect design for annual recirculation, and the recirculation reliably outlives the context.

The predictable result is a January spike of alarmed posts, followed by eleven months in which the figure is quoted without its year. The Clock's own authors are clearer about its limits than most of the coverage: it is intended to start an argument about risk, not to settle a date.

When the setting changes, open the Bulletin's own statement before anything else. It names the specific developments the board weighed. Almost every viral version of the story omits them, which is how a considered risk assessment becomes a screenshot with an alarmed caption.

Then check the year on whatever you are reading. That single step resolves most of the confusion this topic produces.