Jacques Vallée: asking better questions with AI
Source interview: Watch on Rebelliously Curious — Chrissy Newton’s YouTube channel.
Interview analysis
Jacques Vallée uses his early work on astronomical catalogues to explain why the question matters as much as the answer a computer produces. He proposes practical tools for comparing UFO reports and prompting a human witness interviewer, drawing on a longer history of applied AI than recent language systems suggest. Chrissy Newton also asks whether AI might help interpret extraordinary encounter narratives, taking the conversation from information handling into Vallée’s ideas about language, perception and space.
Key themes
Questions the catalogue can answer
Vallée recalls replacing overnight programming with English-language questions about a defined astronomical catalogue. Once the basic system worked, much of his effort went into recognising ambiguity or questions that made no sense for the data. The example of asking about stars with no colour shows why an apparently fluent answer can be the wrong outcome. He wants the system to clarify the question with the user, preserving the connection between a term, its scientific meaning and the records being searched.
A recurring claim becomes a comparison
Vallée takes the claim that UFOs turn blue when accelerating and asks whether the catalogue also contains reports of red light or no light. He describes the large catalogue he helped design as classified. That example gives software a precise role in examining a proposed association, including contrary observations. The catalogue’s descriptions and categories still determine what can be compared. An answer would first concern what witnesses reported, with any physical theory requiring the original observations to be understood.
An assistant beside the interviewer
Vallée imagines a tool that remembers possible explanations and prompts questions an investigator might overlook. His examples concern practical details such as whether an image was always clear or an object disappeared and returned. He reserves the conduct of the interview for the person who can notice hesitation, allow silence and return to a difficult point later. Newton’s questions about empathy lead him to warn that apparent precision may encourage excessive trust, reinforcing the supporting role he assigns to the system.
Recognising an unfamiliar target
A known aircraft offers a reference for classification; an unidentified configuration does not offer the same starting information. Vallée develops that problem through encounter stories involving apparently absurd speech or an interior larger than its exterior. He interprets them through broader ideas about language and the structure of space, including conversations with Eric Davis. These passages extend the discussion beyond retrieving or comparing testimony to a proposed explanation of what the witnesses experienced.
Assessment
Testing the proposed tools
Vallée’s astronomical programme depended on defined questions and a catalogue whose contents could answer them. For future work he also identifies his own unclassified civilian records. A witness assistant or colour–acceleration query could be examined through the reports retrieved and the questions prompted, including mistakes and omitted alternatives. His recollection of an adapting medical robot raises a related problem of validation, which he carries into the harder task of classifying an unknown object. The interpretations of space and language concern the encounters themselves. Better software could help compare those accounts while leaving their physical explanation open.
Navigate the source interview
- 2:00 — Northwestern, astronomy and early computer work
- 5:36 — Natural-language questions about a star catalogue
- 12:30 — Ambiguous questions and answers a system should not give
- 16:46 — UFO questionnaires and adaptive witness interviews
- 19:16 — Vallée’s account of industrial and medical applications
- 23:58 — Risks in automated recognition and decision-making
- 28:38 — Understanding and testing a system’s decisions
- 32:04 — Classifying objects whose properties are unknown
- 35:09 — Why AI and UAP research need not be a new conjunction
- 39:41 — Testing claims about colour and acceleration against a catalogue
- 45:05 — An assistant for the human investigator
- 50:27 — Emotion, apparent understanding and overconfidence
- 56:16 — Encounter narratives, language and interpretations of absurdity
- 67:32 — Practical tools for interviews and coordinated observations
Questions
2:00 What did Vallée learn from developing an early natural-language astronomy system?
16:13 How has he seen AI develop, and where do ambiguous questions cause difficulty?
23:58 What risks arise when a system must recognise unfamiliar objects or support decisions?
39:41 How could catalogues test a recurring claim about reported UFO behaviour?
45:05 Which parts of interviewing a witness should remain with a human investigator?
50:27 Could an apparently responsive system encourage users to trust it too much?
67:32 Which practical applications should researchers develop first?
Selected quotations
Three full stops (...) mark words omitted from the transcript, including verbal filler or false starts. Every retained word follows the source transcript. Use the linked timestamp or full transcript for surrounding context.
if I give you an answer that looks right to a question that wasn't right ... I'm going to lead you into mistakes
Vallée explains why he spent so much time recognising ambiguous or unsuitable questions in his early astronomical system.
there are thousands of cases where people have seen a UFO that accelerated did it always turn blue or were there cases where it didn't have any light were there cases where it turned red
Vallée illustrates a query for supporting and contrary reports using a large case catalogue he helped design and describes as classified.
it's not going to do the interview it cannot do the interview because so much is in the human connection
Vallée assigns the assistant a prompting and memory role while keeping the interviewer responsible for the pace of the conversation and responses to the witness.
Sources
- Source interview on YouTube — Newton’s conversation with Vallée about the history and proposed research uses of artificial intelligence.
Read the full source transcript (plain text). The supplied wording and time markers are retained.