If you have spent any time on social media this week, you may have come across clips from a recent Why Files episode that has sent “global consciousness” trending across timelines. The episode revives a decades-old experiment that claims something extraordinary: that machines built to spit out random numbers seem to behave less randomly at moments when the world feels something together — a terror attack, a tsunami, the death of a beloved figure, the countdown to a new year.
The experiment is real. The data are real. The question is what, if anything, they mean — and that is where the story gets genuinely interesting.
Eggs that listen to the world
The project at the centre of all this is called the Global Consciousness Project, or GCP, sometimes nicknamed the “EGG Project” after the small devices — Electrogaiagrams — that form its backbone. It began in 1998, the brainchild of Roger Nelson, a psychologist who had spent two decades at Princeton University’s Engineering Anomalies Research lab, an outfit better known by its acronym, PEAR, and by its long, controversial history of studying whether human minds can influence machines.
The basic kit is unglamorous: a hardware random number generator, the kind of device that produces strings of zeroes and ones using genuinely unpredictable physical processes such as quantum tunnelling, rather than the pseudo-randomness a computer normally fakes. At its peak, the GCP had roughly 60 to 70 of these devices humming away quietly in homes, offices and university labs across the world — from Princeton to Fiji, from Brazil to Russia — each one generating a fresh 200-bit number every second, and beaming it back to a central archive in real time.
For years, this network simply sat there, generating an ocean of random data, mostly ignored. The interesting part happened only when researchers went looking for it: whenever something happened that captured the attention or emotion of large numbers of people at once — the September 11 attacks, a tsunami, New Year’s Eve, a globally watched funeral — they would mark out a window of time around that event and check whether the random outputs from the network, taken together, deviated from what pure chance would predict.
According to the project’s own published figures, after testing roughly 500 such events over close to two decades, the cumulative statistical departure from chance reached odds of about one in a trillion. In plain terms: if the numbers truly are behaving the way the project says they are, that is not the kind of result you would expect to see by luck.
What the project says this means
Nelson and his collaborators frame this as evidence for what they call a coherent “global consciousness” — the idea that when millions of people share an emotional or attentional state, something about that shared state leaves a measurable trace on the physical world, however faint. Some proponents reach further still, invoking the idea of a “noosphere”, a term coined by the Jesuit philosopher Pierre Teilhard de Chardin to describe an emerging sphere of human thought encircling the planet, much as the biosphere encircles it with life.
It is worth being precise about what is actually being claimed here, because this is where casual coverage often goes astray. The GCP is not claiming that thoughts can bend spoons or stop clocks. The claimed effect, even by the project’s own description, is small — a subtle statistical “structuring” of what should be pure noise, visible only when you pool data from dozens of devices over very large numbers of events and look at the picture in aggregate. Nobody is claiming you could point one of these devices at a single person and read their mind, or that a single dramatic news event produces an obvious, visible spike on its own.
The project has also evolved. Citing what it describes as a catastrophic hosting failure that ended its original data collection on April 3, 2026, the GCP’s founders have begun work on a successor effort, GCP 2.0, which aims to deploy a much larger network — reportedly thousands of random-number channels across dozens of clusters worldwide — partly in response to long-standing criticisms about the statistical power and design of the original network.
The other side of the ledger
And there is a great deal of criticism. The GCP sits squarely within parapsychology, a field that mainstream science treats with deep scepticism, and the project’s claims have been picked apart repeatedly over the years.
The most pointed critique came from Jeffrey Scargle, a NASA astrophysicist, who examined the project’s flagship case study — its analysis of the September 11 attacks — and concluded that the statistical case for an anomalous effect had not been convincingly made. A separate re-analysis by other researchers went further, concluding that the apparent September 11 result looked like a product of how the data were selected and analysed after the fact, and that an alternative method of crunching the same numbers found nothing beyond ordinary chance.
This points to the central methodological worry that critics keep returning to: the GCP is, by its own admission, fundamentally a post-hoc exercise. Researchers decide which events count as significant, decide where to draw the boundaries of the time window around each event, and then check the data against those choices — sometimes long after the fact. Critics argue this leaves enormous room for what statisticians call “researcher degrees of freedom”, where a sufficiently flexible analysis can find a “signal” in almost any sufficiently large dataset, purely by chance, simply because there are so many different ways to slice it.
There is also the question of venue. Much of the GCP’s published work appears in journals such as the Journal of Scientific Exploration — outlets that do conduct peer review, but which sit outside the mainstream scientific literature and are largely dedicated to topics, including parapsychology, that most working scientists do not consider to have a credible evidence base.
None of this means the people involved are acting in bad faith, or that the raw data has been faked — the GCP has long made its archive publicly available, which even sceptical commentators have credited as a mark of transparency. The dispute is almost entirely about interpretation: what counts as a meaningful pattern, and what counts as the ordinary, expected noise of a system that, across enough trials, will throw up apparent “patterns” purely by chance.
Why this is trending now
Part of the answer is simply the Why Files effect. The YouTube channel has a large audience built around exactly this kind of material — real instruments, real institutions, real data, wrapped around a question just tantalising enough to resist a clean yes-or-no answer. That combination travels well online, especially in short-clip form, where the nuance of “small statistical anomaly of disputed significance” easily compresses into “scientists detected the world’s emotions”.
But there is also a broader current here. Questions about consciousness — what it is, whether it can be measured, whether it extends beyond the individual brain — have had a notably high profile over the past year, from high-profile popular science books exploring the frontiers of consciousness research to renewed public interest in meditation, “coherence”, and related ideas from groups such as the HeartMath Institute, which runs its own related Global Coherence Initiative.
Coming at a moment of genuine global anxiety — wars, economic shocks, a sense that the world’s mood is unusually volatile — the idea that there might be some instrument, somewhere, registering that collective unease has an obvious emotional pull, regardless of what the statistics actually support.
The bottom line
The Global Consciousness Project is a real, long-running, methodologically serious attempt to ask a genuinely interesting question, using real hardware and a real, publicly available dataset. Its own analysis claims results that would be remarkable if they hold up. But “if they hold up” is doing a lot of work in that sentence — credible critics within and outside the field have raised substantial, unresolved objections about how those results were obtained, and the project remains firmly outside the scientific mainstream.
For readers encountering this through viral clips this week, the honest summary is this: something unusual shows up in this dataset when researchers go looking for it after major world events. Whether that “something” is evidence of a planetary mind, or simply a vivid illustration of how easy it is to find patterns in noise once you know what you are looking for, remains genuinely, and as yet unresolved, an open question.
