Welcome to the Faultline Observatory

The Faultline Observatory is a permanent public record of important technological claims whose long-term significance remains uncertain. Rather than predicting which technologies will succeed or fail, the Observatory records how evidence changes those claims as it accumulates. Every Frontier Record preserves the history of the claim, including moments when confidence increased, weakened, or changed direction. The goal is not to produce forecasts. It is to create an accountable public record that can be revisited as new evidence emerges.

Many frontier technologies generate excitement long before the evidence is settled. Headlines move on quickly, while the underlying claims continue to evolve over months or years. The Observatory exists to reduce that gap. Instead of following the news cycle, it follows the evidence. Every significant development is assessed in context and added to a permanent record, allowing readers to understand not only what happened, but how confidence has changed over time.

A Frontier Record is the Observatory's core publication. Each Record focuses on a single technological claim — a new quantum computing milestone, an AI capability, a medical breakthrough. The Record documents the original claim, the evidence that follows, and every assessment made by the Observatory. Together these create a transparent history that readers can inspect for themselves.

Many claims do not pass as a Frontier Record. Many claims simply don't naturally generate the kind of long-lived evidence pressure that makes a Frontier Record valuable. Many claims are scientifically interesting and notable discoveries. But they're simply not part of our set of frontier programmes—where claims evolve over years under sustained technical uncertainty and institutional significance. Even if another paper appears on a claim, it’s likely to be incremental rather than a continuing public claim whose assessment people will want to revisit over a decade. One of the strengths of FCIF is that it doesn’t try to become a catalogue of every fascinating scientific result. Saying “no” to good science is part of maintaining a coherent corpus.

A Frontier Record neets to meet specific criteria. A claim is admitted when it satisfies three conditions: it makes a specific, checkable assertion rather than a general trend; the evidence available today is insufficient to resolve it; and further evidence — a publication, a replication attempt, a deployment, a retraction — can realistically be expected to arrive. The objective is not to document every news story, but to build a durable public record of developments that matter.

A Frontier Record is designed to evolve. As new evidence becomes available, the Observatory publishes additional assessments explaining whether that evidence strengthens, weakens, or meaningfully changes the claim. Earlier assessments are never rewritten or removed. Instead, the Record preserves the complete trajectory, allowing readers to see how understanding developed over time. The Observatory preserves trajectories rather than verdicts. Understanding changes, but the record of how it changed remains.

Every Frontier Record carries the Observatory's current assessment of the evidence. As new evidence appears, that assessment may change. The current assessment is a snapshot of where the claim stands today. The assessment history — preserved in full, never edited away — shows how that judgement evolved to reach it.

The specific assessment states, and how they are assigned, are explained in the Methodology.

Four is the current number, not a fixed one. The four Programmes — Quantum Engineering, Artificial Intelligence, Advanced Materials, and Biotechnology — emerged from the claims the Observatory actually took on, each representing a distinct type of contested frontier discovered through the founding records rather than assigned in advance. A new Programme is added only when a genuine, recurring frontier appears that does not fit within the existing four — not simply because a new claim seems interesting on its own. That threshold is intentionally high: the Observatory grows by demonstrated pattern, not by anticipation.

The Observatory is organised around the Frontier Record. Every Record belongs to a Programme. Each Programme includes a Reading Room, which brings its Records together in one place, alongside Programme Notes on the institution tracking that field, and Landscape Essays — longer-form analysis of the frontier itself, independent of any single claim. The Observatory is built from five connected parts: Frontier Records, Programmes, Reading Rooms, Programme Notes, and Landscape Essays. Together they create a single public record, viewed at different levels of detail.

Trust should come from transparency rather than authority. The Observatory evaluates evidence using a published methodology rather than personal opinion. Every assessment considers the quality of the evidence, the reliability of its source, and its significance for the underlying claim. That reasoning is recorded publicly, alongside the linked evidence and the assessment history, and previous conclusions remain visible even when later evidence changes them. This lets readers examine not only what the Observatory concluded, but why — and not only the current assessment, but how it was reached.

The Faultline Observatory is an independent public-interest project, founded and maintained by Stuart Kegel. Its credibility is intended to rest not on its founder, but on the transparency of its methods, the permanence of its public record, and the accountability of every published assessment.

You can read how the project began in Origins.

There are several ways to continue exploring the Observatory.

You can browse Frontier Records to follow individual claims, visit a Programme's Reading Room to see developments across a field, or read a Landscape Essay for longer-form analysis of a broader theme.

However you choose to explore, every part of the Observatory is connected by the same principle: tracking technological claims until the evidence decides.