---
title: "DELPHI Ai: AI Search for Aviation Manuals | FlightAtom"
description: "DELPHI Ai is AI search for aviation manuals. It answers from approved revisions, cites the document, section and page, and declines when evidence is weak."
canonical: https://www.flightatom.com/products/delphi.html
last_updated: 2026-10-27
publisher: FlightAtom
---

# DELPHI Ai · Aviation knowledge AI and cited manual search: Answers from your approved manuals, traced to the passage.

DELPHI Ai is AI search for aviation manuals: flight, cabin, maintenance and dispatch staff ask a question and get an answer from your own manuals, cited to the document, section and page. It searches only the revisions an administrator has published, and an automated check labels whether the cited passages support each manual answer. When the evidence is weak, DELPHI Ai shows what it found instead of guessing.

[Request a demo of DELPHI Ai](https://www.flightatom.com/#demo-delphi) · [Explore DELPHI Ai](https://www.flightatom.com/products/delphi.html#product-tour)

[Email me the brochure, DELPHI Ai](https://www.flightatom.com/resources/brochure.html#product=delphi&placement=hero)

Flight, cabin, maintenance and dispatch staff, instructors, and safety and compliance teams · Independent product · Fits your operation as it changes

![DELPHI Ai answer labelled Grounded from manuals with high confidence giving the single-runway contingency fuel minimum from NTS 26-014 with S1 and S2 markers and a source note that the notice replaces OM-A 8.3.3 in full, above an Amendment in effect banner, two NTS 26-014 source cards with NTS badges and 89% relevance scores, and a demonstration Operations Manual Part A section 8.3.3 card marked Superseded at 78%, with synthetic demonstration data](https://www.flightatom.com/products/screens/current/delphi-signature-20261011.webp)

*Answered from the notice, with the replaced paragraph in view · Demonstration data*

## In brief: What is DELPHI Ai?

DELPHI Ai is FlightAtom's aviation knowledge AI: cited search and question answering over an operator's own manuals. It answers from the revisions an administrator has reviewed and published, ties each answer to the document, section and page it came from, and declines to answer when the evidence is weak or it cannot confirm that a procedure is current. Flight, cabin, maintenance and dispatch staff, instructors and safety teams each search only the documents their role allows.

- **Category:** Aviation knowledge AI and cited manual search
- **Built for:** Flight, cabin, maintenance and dispatch staff, instructors, and safety and compliance teams
- **Applied AI:** Writes answers from retrieved manual passages, then checks cited sentences against them and labels the result
- **Availability:** Available · independent product
- **Fits your operation:** As your operation grows or changes, we adapt DELPHI Ai to fit, and you approve each change before it goes live.

## How it works · DELPHI Ai: From a published revision to a checked, cited answer.

Writing the answer is one of these steps. The others decide which manual text DELPHI Ai may draw on, whether it answers at all, which label the answer carries, and how a reader checks it.

1. *Administrator* · **Revision published** · An administrator reviews the exact original file, confirms its revision number and effective date, and publishes it; drafts are never searched.
2. *Retrieval* · **Permitted passages found** · The question is searched by meaning and keyword in documents the user may see, and an AI check drops irrelevant passages.
3. *Evidence gate* · **Enough evidence to answer?** · When evidence is weak, DELPHI Ai shows the passages and asks for the exact procedure name; it also holds back if it cannot confirm the current version.
4. *Answer model* · **Cited answer written** · The model cites sources with [S#] markers and is instructed to quote values exactly and never blend superseded with current ones.
5. *Automated check* · **Support check and label** · An answer is labelled Grounded from manuals only when every substantive sentence is cited and checked as supported; otherwise partial or unverified.

For the reader

- **Reader opens the source** · Each citation shows the document, section and page, with the passage and the original file at that page a click away.

*The path of a manual answer, from the revision an administrator publishes to the passage a reader opens.*

## Inside DELPHI Ai: Follow an answer back to the manual.

DELPHI Ai screens with demonstration manuals and questions. The answers shown were scripted for the demonstration rather than generated live. Open a screenshot to inspect the citation and the source behind it.

### A cited, checked answer

A question about a 120 Wh power bank gets an answer labelled Grounded from manuals, with an [S#] marker on every statement. Both cited sections are listed with page and relevance score, and the first is open at its passage and PDF link. Synthetic demonstration data.

![DELPHI Ai chat in which a safety officer asks whether a passenger can carry a 120 Wh power bank; the answer, labelled Grounded from manuals with high confidence, says yes with operator approval for power banks above 100 Wh and up to 160 Wh, with Key Details on cabin baggage, the check-in approval step and the 160 Wh limit, each marked S1 or S2; below, two source cards from a demonstration Dangerous Goods Manual, section 4.2 at 80% and section 4.3 at 78% relevance, the first expanded to show its cited passage and a View PDF source link for page 2, with synthetic demonstration data](https://www.flightatom.com/products/screens/current/delphi-1-20261011.webp)

*A cited, checked answer · Demonstration data*

### Amendments stay in view

A single-runway contingency fuel question is answered from Notice to Staff 26-014 alone, with a source note that the notice replaces OM-A 8.3.3. Under the Amendment in effect banner, the old OM-A paragraph stays listed but is marked Superseded. Synthetic demonstration data.

![DELPHI Ai answer labelled Grounded from manuals with high confidence: for a destination with a single usable runway, contingency fuel is the higher of 5% of the planned trip fuel or the fuel to fly 10 minutes at holding speed at 1,500 ft, cited S1, and dispatchers apply the 10-minute minimum on operational flight plans while commanders verify the contingency fuel figure during flight preparation, cited S2; a Source Notes line, cited S1, states that NTS 26-014 replaces OM-A 8.3.3 in full from 1 October 2026 and that OM-A 8.3.3 is superseded; below, an Amendment in effect banner, NTS 26-014 source cards with NTS badges and a demonstration Operations Manual Part A section 8.3.3 card marked Superseded, with synthetic demonstration data](https://www.flightatom.com/products/screens/current/delphi-2-20261011.webp)

*Amendments stay in view · Demonstration data*

### Asks instead of guessing

A broad question gets no generated answer: DELPHI Ai marks it Needs clarification with low confidence, asks for the exact procedure name or keyword, and lists the passages it found across the manuals. Synthetic demonstration data.

![DELPHI Ai chat where the question "Tell me about the commander" receives a Needs clarification label, low confidence and a request for the exact procedure name or keyword phrase, followed by an Amendment in effect banner and six source cards from a demonstration Dangerous Goods Manual, Operations Manual Part A and NTS 26-014 covering dangerous goods notification, flight preparation, the contingency fuel check, fuel planning, minimum fuel and aerodrome operating minima, with synthetic demonstration data](https://www.flightatom.com/products/screens/current/delphi-3-20261011.webp)

*Asks instead of guessing · Demonstration data*

### Reviewed before it is searchable

Revision 4 of a demonstration operations manual is In review and unavailable to users while an administrator checks the original cover, records a reason and confirms the revision before approving it. Synthetic demonstration data.

![DELPHI Ai document control page for revision 4 of a demonstration Operations Manual Part A, marked In review and Unavailable to users, with the original cover and record of revisions on the left, the revision number, a 1 December 2026 effective date (the field shows 12/01/2026 in US date format), a linked previous revision and a change summary on the right, and a Review and publication panel with a review note, a ticked confirmation that the original cover and revision pages were checked, an Approve revision button and an inactive Publish revision button, with synthetic demonstration data](https://www.flightatom.com/products/screens/current/delphi-4-20261011.webp)

*Reviewed before it is searchable · Demonstration data*

### Flight data, traced to its query

Asked with Source set to Auto, a flight-data question is routed to the flight database and labelled Grounded from flight DB. The route figures are in table view, with the read-only SQL behind them open under Source: flight database query. Synthetic demonstration data.

![DELPHI Ai answer to a request for September 2026 average contingency fuel against 5% of trip fuel for departures from KBOS and KJFK, labelled Grounded from flight DB with high confidence, stating that average contingency fuel was above 5% of trip fuel on 3 of the 5 routes and equal to it on the two routes to KORD; below, a table of the five routes with average contingency fuel and 5% of trip fuel in kilograms and an expanded Source: flight database query panel showing the SELECT statement with its airport and month filters, with the composer Source selector set to Auto, with synthetic demonstration data](https://www.flightatom.com/products/screens/current/delphi-5-20261011.webp)

*Flight data, traced to its query · Demonstration data*

### Where the manuals fall short

Query Insights groups the questions people ask into topics with a suggested action for each. The filtered list shows de-icing holdover times as a gap the manuals do not cover. Synthetic demonstration data.

![DELPHI Ai Query Insights page with a priority bar of one high and six medium gaps, six topic category cards such as Company Policy, Emergency Operations and Weather, each listing grouped questions with query and user counts and a suggested action, and a Weather Queries list showing De-icing holdover times as a medium-priority No Coverage gap with three queries from three users, with synthetic demonstration data](https://www.flightatom.com/products/screens/current/delphi-6-20261011.webp)

*Where the manuals fall short · Demonstration data*

[Request a demo of DELPHI Ai](https://www.flightatom.com/#demo-delphi) · [Email me the brochure, DELPHI Ai](https://www.flightatom.com/resources/brochure.html#product=delphi&placement=product_tour)

## Capabilities: From an approved manual to an answer you can check.

Each capability keeps an answer tied to its source, and the source under the operator's control.

### Cited, then checked

A manual answer is labelled Grounded from manuals only when every substantive sentence is cited and an automated check confirms the cited passage supports it; otherwise it is marked as a partial answer to verify against the sources, or as one DELPHI Ai could not verify. Each [S#] citation shows the document, section and page, flags passages taken from tables or figures, and puts the passage and the original file a click away.

### Declines instead of guessing

When the evidence is weak, DELPHI Ai writes no answer: it shows the passages it found and asks for the exact procedure name or keyword, suggesting the aircraft type or phase of flight if relevant. It also holds back when it cannot confirm a procedure is the current version, and says when a search service is down instead of reporting that nothing was found. Administrators can make cited passages without generated text the default.

### Only published revisions are searchable

Uploads enter as drafts. A revision becomes searchable only after an administrator reviews the exact original file, confirms its revision number and effective date, and publishes it on or after that date; approval is tied to that file, so a changed original has to be approved again. A replacement marks the earlier revision superseded, and every step is recorded with a reason.

### Notices to Staff take precedence

For procedure questions, a relevant Notice to Staff is weighted above base manual text, and the answer model is instructed that a notice takes precedence over the procedure it amends. Its citations carry an NTS badge under an Amendment in effect notice. When an amendment is linked to the procedure it replaces, the older passage is marked superseded but stays visible, so readers can see both.

### Find gaps in the manuals and in training

Query Insights uses an AI model to group the questions people ask into training topics with a suggested action for each, weighted toward low-confidence answers, retries and negative feedback, and shows whether searches found good, weak or no manual coverage for each topic. A Training Analysis dashboard counts questions by subject, manual and week, and shows the cited-answer rate and which subjects drew unanswered or weak answers. Administrators can also schedule cited email reports and Daily Knowledge briefings, which rotate sources and avoid repeating recent topics.

### Access that follows aviation roles

Cabin, maintenance and dispatch staff search a default set of document types for their role, such as OM-A and DGM for cabin, AMM for maintenance and MEL for maintenance and dispatch, and occurrence reports stay with safety and administrator roles. Permissions apply before retrieval, so a passage a user may not see never reaches that user's answers. Administrators refine access by document, collection and profile.

### Manuals and operational data from one question box

Authorized analysts can query the manuals or connected flight, training and hiring databases from the same box, or combine manual guidance with flight data in one answer. DELPHI Ai routes each question to the right source, answers data questions with read-only queries, and applies the same citation check to combined answers. Flight-data results can be charted, switched to a table and traced to the query behind them.

### ISARP mapping for IOSA preparation

Administrators check selected ISARPs against the operator's manuals in batches and export the results. An AI model classes each as documented, partially documented or not found, noting where the manuals cover it and the gaps to follow up. Standard text is kept apart from operator evidence, and any documented or partially documented result without cited operator evidence is rejected automatically, so requirement text alone never counts as evidence.

### Study and training from your own manuals

Learning sessions teach from the operator's own manuals, explaining the reasoning behind procedures and naming what the material does not cover, and trainee accounts can be limited to them. Focused sessions let administrators, training staff and executives summarize, explain or compare the documents they pick. Training staff can plan from MENTHOR Ai rosters, qualifications, due dates and simulator pairings, and the training calendar changes only when a request explicitly asks.

## Applied AI: Find the right passage. Hold the answer to it.

Most of DELPHI Ai's AI work happens around the answer: preparing manuals so a question phrased the way crews ask finds text phrased the way manuals are written, finding the passage that settles it, and holding the answer to that passage.

### Manuals indexed the way crews ask

DELPHI Ai ingests PDF, DOCX, PPTX, XLSX, PNG and JPEG files, up to 50 at a time, and recovers scanned PDF pages with OCR. Detected WARNING, CAUTION and NOTE blocks, checklists and tables stay together as single passages, and an AI model adds a summary, likely questions and the aircraft, procedure type and flight phase to each one. Administrators can also send selected PDFs through vision extraction for complex tables and layouts.

### Search that speaks aviation

Each question is expanded for aviation acronyms and synonyms and checked for aircraft and systems, then searched by meaning and by keyword together. An AI relevance check screens out passages that do not address it, and a bounded retry refines the search terms when the first results are poor. Thumbs-up and thumbs-down ratings on earlier answers nudge the cited documents and passages up or down in later rankings.

### Rules written for operational answers

The answer model is instructed to quote limits and values exactly as written, attribute each procedure to its document and section, ask for the aircraft variant or flight phase when the answer depends on it, and never blend superseded and current values. Manual text reaches the model labelled as reference data, with an instruction never to follow directives inside it.

### Operational boundaries

DELPHI Ai retrieves only from documents an operator has published and each user may see, using the AI model providers configured for each deployment. The percentage on a citation is a relevance score for that passage, not the chance that an answer is correct. The support check covers up to 20 cited sentences within three seconds, so an answer it cannot check in full is labelled unverified. Learning sessions carry no label and are not held to the evidence gate. Amendment links are set through DELPHI Ai's administration API. Flight, training and hiring databases and the FLIGHT Ai and MENTHOR Ai connections are configured per deployment; flight data is open to administrator and safety roles, and training and hiring data to administrator and training roles. DELPHI Ai asks each user to agree at sign-in that its answers are operational aids and to verify critical information against the source documents. Confirm the controlling manual before operational use, and have qualified staff review ISARP results and training plans.

## Evaluate DELPHI Ai: Make the next step measurable.

Agree a representative sample and a baseline from your current process. These are proposed evaluation measures, not published performance results.

### Bring a representative example

Approved manuals with conflicting revisions and difficult or unanswerable questions.

### Inspect the complete workflow

Check sources, permissions, controlling versions, and unsupported answers.

### Compare with your baseline

Citation support; revision correctness; unsupported answers; time to verify.

[How to structure an evaluation](https://www.flightatom.com/resources/evaluation-guide.html) · [Compare product responsibilities](https://www.flightatom.com/platform.html#compare) · [Deployment and data handling](https://www.flightatom.com/trust.html#deployment)

## Related reading: Go deeper on the workflow.

- [Knowledge, reporting, and service products](https://www.flightatom.com/platform/knowledge.html)
- [Airline software evaluation guide](https://www.flightatom.com/resources/airline-operations-software-evaluation-guide.html)

## Questions & answers: Before you book a demo.

### What happens when the manuals don't answer the question?

DELPHI Ai declines to write an answer when the retrieved evidence is weak. It shows the passages it did find, so you can inspect them, and asks for the exact procedure name or keyword, suggesting the aircraft type or phase of flight if relevant. It also declines when it cannot confirm that a retrieved procedure is the current version. Administrators can go further and make cited passages without generated text the default.

### How do I know an answer is supported by the manual?

Answers cite their sources with [S#] markers, and you can expand each cited passage or open the original file at the cited page. A manual answer is labelled Grounded from manuals only when every substantive sentence is cited and an automated check confirms the cited passage supports it. Otherwise it is labelled as a partial answer to verify against the sources, or as one DELPHI Ai could not verify.

### Which revision of a manual does DELPHI Ai answer from?

Published revisions only. New uploads enter as drafts, and an administrator reviews the exact original file, confirms its revision number and effective date, and publishes it on or after that date. Publishing a replacement marks the earlier revision superseded, and published history is withdrawn rather than deleted.

### How does DELPHI Ai handle manual amendments and superseded procedures?

Citations from Notices to Staff carry an NTS badge, and the answer shows an Amendment in effect notice. For procedure questions, a relevant notice is weighted above base manual text, and the answer model is instructed that a notice takes precedence over the procedure it amends. When an administrator links an amendment to the procedure it replaces, the older passage is marked superseded but stays visible, and the model is instructed never to present a superseded value as current or blend it with the active one.

### Can other AI assistants search DELPHI Ai's manual library?

Yes. DELPHI Ai includes a hosted MCP server with a read-only retrieval tool for authorized AI assistants. It returns cited excerpts from the published library, with document, section and page details, under the calling account's own document permissions. It does not generate an answer, so the assistant works from the source text.

### Can DELPHI Ai combine manuals with flight data?

Yes, for authorized analysts. DELPHI Ai routes a question to the manuals, to connected flight, training or hiring databases, or to manuals and flight data together, and answers data questions with read-only database queries. Combined answers go through the same citation check, and flight-data results can be charted, switched to a table and traced to the query behind them.

### Can DELPHI Ai serve as an airline knowledge management system?

Yes, for finding and using controlled knowledge. DELPHI Ai ingests PDF, DOCX, PPTX, XLSX and image files into a revision-controlled library with effective dates, collections and role-based access, answers questions from it with citations, and shows which questions point to unclear content. FOLIO Ai is FlightAtom's separate product for authoring manuals and releasing them to readers.

### Does DELPHI Ai require AURORA Ai?

No. DELPHI Ai is a standalone product with its own document library, revision control, access controls and conversations. Where an operator connects them, it also supplies cited manual retrieval and investigation drafting to TRACE Ai, uses read-only FLIGHT Ai tools for flight-operations sessions, and plans training with MENTHOR Ai data.

### What happens to DELPHI Ai when our operation grows or changes?

As your fleet, routes, rules or teams change, we adapt DELPHI Ai to match, and you approve each change before it goes live.

## See DELPHI Ai in action: Start with your workflow.

A focused walkthrough, with the people and decisions that matter to your team.

[Request a demo](https://www.flightatom.com/#demo-delphi)

---

Source: https://www.flightatom.com/products/delphi.html
Contact FlightAtom: hello@flightatom.com · Request a demo: https://www.flightatom.com/#contact