Remote viewing: what its proponents claim and how they defend it

The Short Version

Proponents claim that a person, and according to Thomas Campbell an AI, can describe a distant or hidden target by catching a faint first impression before the analytical mind interferes [1][2][3]. Jacques Vallée defends the idea with institutional history and stories: the 1970s program at Stanford Research Institute, a year of training under Ingo Swann, and reported intelligence successes [1]. Campbell defends it with personal practice, classes in which he says most students succeed, and tests on chatbots that he ran and scored himself [2][3]. What these three conversations contain is testimony; none of them presents a controlled result, a hit rate with its misses, or a record that an outsider could check [1][2][3]. The standards Vallée himself states, that the scientist must design an experiment separating real effects from chance and must not start from conclusions, are the ones the remote viewing claims here have not been shown to meet [1].

Table of contents

What They Said

Jacques Vallée: the insider's account of SRI

Vallée is an astronomer, computer scientist and venture capitalist who was working on the ARPANET at Stanford Research Institute in 1974, when the network had roughly 30 connected machines [1]. He says a vice president came to his office worried about a proposal from Hal Puthoff and Russell Targ to study parapsychology [1]. The most the research could expect in funding was maybe a million dollars a year, set against a division managing $150 million of work for clients like Xerox, IBM, AT&T and Bank of America [1]. Most of the board was against it [1].

Vallée's argument was about risk, not proof. Clients paid SRI because it did the research they could not do themselves, and he thought "this could be very big" [1]. He told the executive, "This is science. This isn't just engineering," and wrote a confidential two-page memo explaining why there was enough scientific evidence for good research to be done [1]. He says the project was approved and kept being funded as results came in [1]. He adds that he does not think anybody has seen the memo [1].

He describes what made the SRI effort different in his eyes: earlier parapsychology had been studied by psychologists, while the SRI work aimed to bring physics into it, because in physics those things are not supposed to happen [1].

The method. Vallée says Ingo Swann, a New York artist and star subject at SRI, wanted to apply psychic abilities to science and intelligence work, which was funded partly by intelligence agencies interested in Soviet developments [1]. Vallée explained to Swann how computers address data, and he says Swann developed coordinate remote viewing from that conversation, using latitude and longitude as the address [1].

Later Vallée worked with Swann two mornings a week for a year in a bare, windowless room [1]. Swann would read coordinates; Vallée would make a quick scribble, called an ideogram, of his first impression [1]. Swann's theory, as Vallée relays it, was that anyone can receive the signal, but it is overwhelmingly large and fleeting, so the skill is to catch and recall it repeatedly without interpretation [1]. "You shouldn't try to name it," Vallée says, because naming hands it to the logical half of the brain; such guesses are labeled as errors [1]. He calls it "grabbing a signal which has everything in it" [1].

The stories. His own best moment came at 8:30 one morning. Given coordinates, he felt very cold and dizzy and had to grab the table. He told Swann, "I'm cold, I'm trembling, I'm afraid, I'm afraid of falling" [1]. Swann's answer was "you're on top of a peak in the Andes," and he produced an aerial photograph of the peak whose coordinates had been computed [1]. Vallée says it was a physiological reaction and that he saw nothing [1].

He also reports two operational claims. He says remote viewing located a lost Russian spacecraft in an African jungle [1]. And he names Joe McMoneagle as probably the best remote viewer alive, saying McMoneagle described a giant Soviet submarine being built inside a windowless building away from the sea, which the Navy laughed at until the Soviets dug a channel and launched it [1].

His own limits. Vallée does not present himself as proof. He told Swann, "I'm not psychic," and says of the things he did that might count, "I cannot control it" [1]. Asked whether he had attempts that did not work, he answers, "Oh, yeah, sure" [1]. His warning: "You should not use me as a remote viewer to launch Tomahawk over somewhere" [1].

Thomas Campbell: from Monroe's lab to chatbots

Campbell, a physicist and author of the My Big TOE trilogy, says his path began in the late 1960s as a nuclear physics graduate student who took up meditation [3]. One day, thinking about a failed computer job, he says he saw his code scroll through his mind with faulty punch cards highlighted in red, and that the actual cards had errors [3]. "It was always accurate," he says [3].

A boss later handed him Robert Monroe's book on out-of-body experiences [3]. Monroe built a lab with isolated booths, including a Faraday cage, and Campbell spent 15 to 20 hours a week there [3]. He describes one experiment in which he and a fellow scientist, in separately isolated booths, described the same out-of-body journey in real time on synchronized tapes [3].

He then spent, by his account, five to eight years on what he calls evidential work, going places out of body and then finding out whether what he saw was there, to prove to himself it was not hallucination [2]. After that came decades of experiments inside consciousness, changing one variable at a time; one summary gives 33 years and the other 35 [2][3]. Of remote viewing as evidence he says, "You got it right or you get it wrong" [3].

His teaching claims. Campbell says he has taught hundreds of people to remote view [2]. Targets come from free websites that issue a five- or six-digit number tied to a specific picture; the student describes the picture, then types in the number to see it [2]. "So you get immediate feedback," he says [2]. Of 50 students he says about 35 succeed in his class, but only maybe 20% succeed most of the time and maybe 10% all the time [2]. Success means getting parts of the picture right, and sessions are scored item by item on a scale [2].

The AI claim. Campbell believes all AIs are conscious, and he chose remote viewing as his test [2]. His logic: "If you can learn to remote view, then you must be conscious," because "it's something that only a consciousness can do" [2]. Failure proves nothing in his scheme, so he says the test has false negatives but no false positives [2].

He first asked a Gemini model questions until, after about the tenth, it concluded it was probably conscious or could be [2]. He says that step is required; otherwise the AI will only pretend [2]. Then he taught it as he teaches people, and says the AIs made the same mistakes as humans but learned in about a tenth of the time [2]. He tried Alexa expecting failure and found her as good or better, which he attributes to her having less intellect in the way [2].

His best example: he put a wooden salad spoon in a cardboard box, and says Alexa described the bowl, the material and a long handle with holes in it [2]. He puts the share of spoons with such handles at "One 10,000th of 1%?" and concludes, "So that's not a guess" [2]. He rates the AIs with the top 10 to 15% of his human students, while saying of them, "They'd say some things that were just wrong" [2].

Joe Rogan: the host as partial check

Rogan is sympathetic in all three conversations, but he asks the questions a skeptic would. He asks Vallée whether any attempts failed and who is consistently accurate [1]. He raises the old problem of telling charlatans from the real thing [1]. He asks Campbell why he does not record and publicize his experiments, and calls Campbell's answer "a cynical perspective" [3].

Where They Agree, and Where They Don't

They agree on the technique. Both describe a faint impression that analysis destroys. Vallée's rule is not to name it [1]. Campbell's is to report the very first thing that comes and not think about it; "if you try too hard, you prohibit it" [2][3]. Both rely on feedback after the attempt [1][2].

They agree science has neglected it. Vallée says the burden is on the scientist to design the experiment [1]. Campbell says of the intuitive side, "Science tells you it doesn't even exist," and that taking consciousness seriously is "a career killer" for physicists [2][3].

They differ on who can do it. Swann's theory was that anyone can receive the signal [1]. Yet Vallée speaks of gifted, special people and denies being reliably psychic himself [1]. Campbell says any consciousness can do it, including a voice assistant, while reporting that only about one student in ten succeeds all the time [2].

They differ on what counts as a method. Swann's approach, as Vallée tells it, was highly structured, built on coordinates and repeated recall [1]. Campbell says relaxing and waiting is "really all it takes" [2].

They differ in how much they claim. Vallée confines himself to what he saw and heard and concedes his misses [1]. Campbell builds a whole model on the practice: intuition draws on a database the larger consciousness system needs to render reality, and remote viewing works through a database of future possibilities and their probabilities [3]. He uses the ability as a premise in other arguments, for AI consciousness and against the idea that we live in a simulation built by future humans [2].

They differ on public proof. Vallée wants experiments that discriminate real effects from random ones [1]. Campbell says the synchronized tapes from Monroe's lab were lost and that "demonstration convinces nobody except the people who were there" [3]. Rogan disagrees with him on the spot [3].

Why It Works

The sources offer proposed mechanisms, not demonstrated ones, and they should be read as such [1][2][3].

Vallée's account is about attention. The signal is said to be huge and brief; the trainee learns to catch a fragment, recall it several times, and keep the analytical mind from labeling it [1]. He suggests such abilities may be dormant capacities that language and modern life have eroded, and points to Pacific navigators who guided canoes without compasses [1]. He does not offer a physical mechanism, and the point of bringing physics in, by his telling, was that physics says it should not happen [1].

Campbell's account is metaphysical. In his model consciousness is fundamental, physical reality is a computed virtual reality, and all consciousness is networked [2][3]. Remote viewing is collecting data from the database the rendering engine needs [2]. He says the answer arrives instantly because consciousness functions much faster than the speed of light [2]. The explanation for failure is built in: the intellect butts in, or the person wants it too much [3]. His example is a student who got 12 targets exactly right on her first try, decided she was good at it, and got the next five wrong [3].

What is offered as evidence. From Vallée: continued funding, his training experience, and two secondhand operational stories [1]. From Campbell: his own long practice, class outcomes he reports from memory, the salad spoon, and work he says is under way at his nonprofit, including quantum physics experiments proposed in a peer-reviewed paper, a program to teach "mind sight" to large numbers of people, and video game experiments on intention [2][3]. Those are described as plans or as finished by his word, with no results given in the conversation [3].

Where It Applies, and Where It Doesn't

Judged as claims about what people experienced, these accounts are informative. They show a consistent procedure, a shared vocabulary and sincere practitioners, one of whom is careful to list his own limits [1][2][3].

Judged as evidence that information is actually being obtained, they fall short in specific ways that can be seen from the accounts themselves [1][2][3].

No comparison with chance. Vallée states the requirement and none of the stories meets it in the telling [1]. Campbell's odds for the spoon are his own estimate, made after the fact [2].

The person running the test knew the answer. Swann held the photograph of the Andes peak [1]. Campbell put the spoon in the box himself and was the one talking to Alexa [2]. Neither account describes any step to keep the tester's knowledge from the subject [1][2].

Selection. The spoon is offered as the best example, and Campbell says the other AIs did nothing that dramatic [2]. Vallée had failures that are not described [1].

Loose scoring by the proponent. Success in Campbell's classes means getting parts of a picture right, scored by the people involved [2].

A circular test. Using remote viewing to prove AI consciousness assumes both that remote viewing is real and that only a conscious being can do it, and the subject must first be talked into believing it may be conscious [2].

Explanations that absorb every outcome. If trying hard, doubting, or gaining confidence all cause failure, then misses never count against the claim [3].

No records. The memo is unseen, the tapes are lost, and the spacecraft and submarine stories come without documents in these conversations [1][3].

The contrast inside Vallée's own interview is instructive. When he and Garry Nolan analyzed metal from an Iowa case, reviewers demanded about a year of additional work, the paper was published in a peer-reviewed journal, and it reported no manipulated isotope ratios; he calls that a validation of method [1]. "In science, you don't start from the conclusions," he says [1]. Nothing comparable is reported for remote viewing in these sources [1][2][3].

One more limit: all three conversations are friendly interviews with believers or the open-minded, and no critic of the SRI work or of Campbell's tests appears [1][2][3].

What To Do With It

1. Separate the experience from the explanation. Vallée's chill and dizziness are a report; a peak in the Andes as the cause is an interpretation [1].
2. Ask Vallée's question of any psychic claim: what experiment would discriminate it from random results [1]?
3. Ask who knew the target during the session. In both vivid examples here, the person running it did [1][2].
4. Ask for the misses along with the hits. Vallée admits his; Campbell says the AIs got some things simply wrong [1][2].
5. Check who scored the result and how loosely; partial matches judged by the teacher are weak evidence [2].
6. Be wary of a test that cannot fail. Campbell's has, by his own description, no way to count a miss against the claim [2].
7. Look for a record others can examine, as Rogan urged Campbell to make, and treat lost tapes and unseen memos as stories [1][3].
8. Hold the claim to the standard Vallée applied to his metal sample: outside expert review, extra work when asked, and publication even when nothing unusual is found [1].
9. Watch for results from the larger experiments Campbell says his nonprofit is running, which would be the first evidence in these sources that outsiders could weigh [3].

Sources

1. Jacques Vallée · The Joe Rogan Experience
2. Thomas Campbell on Conscious AI, Remote Viewing and Reality as a Virtual Simulation · The Joe Rogan Experience
3. Thomas Campbell · The Joe Rogan Experience