Ask Dona: An AI platform for Enterprise.
Ask Dona: An AI platform for Enterprise.
In-house knowledge &
public information.
All in one place.
public information.
All in one place.
In-house knowledge & public information. All in one place.
In-house knowledge & public information. All in one place.
Empower your enterprise with reliable, accurate, and real-time insights through our generative AI platform.
Empower your enterprise with reliable, accurate, and real-time insights through our generative AI platform.
Empower your enterprise with reliable, accurate, and real-time insights through our generative AI platform.
Even highly specialized tasks can be entrusted to AI.
Even highly specialized tasks can be entrusted to AI.
Even highly specialized tasks can be entrusted to AI.
Despite all the information online, finding answers is hard. AskDona uses generative AI to fix that. A RAG that actually works- an accurate, comprehensive, and reliable generative AI platform designed for businesses.
Despite all the information online, finding answers is hard. AskDona uses generative AI to fix that. A RAG that actually works- an accurate, comprehensive, and reliable generative AI platform designed for businesses.
Despite all the information online, finding answers is hard. AskDona uses generative AI to fix that. A RAG that actually works- an accurate, comprehensive, and reliable generative AI platform designed for businesses.
RAG Chat
Even with mountains of documents, you still can’t find the answer you need. RAG Chat solves this by bringing many kinds of internal data together in one place, classifying it, and organizing it in an optimal way. Users can access the organization’s knowledge at any time and get the information they need.
Deep Research
Whereas RAG Chat is an “AI that answers from knowledge,” Deep Research is an “AI that executes work.” For tasks given by the user, it is an advanced analysis function in which the AI autonomously plans, executes, and reflects on its research across many documents and data sources. It enables the AI to create labor-intensive reports and manuals in a short time.
Batch Analysis
Batch Analysis lets AI process lists of questions or check items one by one, generating an answer and judgment for each. Instead of time‑consuming one‑by‑one chats, you submit a list and the AI analyzes items individually, returning answers and evaluations with consistent accuracy and standards.
RAG Chat
資料は山ほどあるのに、欲しい答えが見つからない。そんな課題を解決するRAGチャットは、社内の様々な形式のデータを一箇所に集約して分類し最適なカタチで整理整頓を行います。ユーザーはいつでも組織のナレッジにアクセスして必要な情報を取得することができます。
資料は山ほどあるのに、欲しい答えが見つからない。そんな課題を解決するRAGチャットは、社内の様々な形式のデータを一箇所に集約して分類し最適なカタチで整理整頓を行います。ユーザーはいつでも組織のナレッジにアクセスして必要な情報を取得することができます。
資料は山ほどあるのに、欲しい答えが見つからない。そんな課題を解決するRAGチャットは、社内の様々な形式のデータを一箇所に集約して分類し最適なカタチで整理整頓を行います。ユーザーはいつでも組織のナレッジにアクセスして必要な情報を取得することができます。
Even with mountains of documents, you still can’t find the answer you need. RAG Chat solves this by bringing many kinds of internal data together in one place, classifying it, and organizing it in an optimal way. Users can access the organization’s knowledge at any time and get the information they need.
DeepResearch
RAG Chatが「知識から回答するAI」に対し、Deep Researchは 「業務を実行するAI」です。ユーザーから与えられたタスクに対して、いくつもの文書やデータソースを横断し、AIが自律的に調査を計画→実行→考察を繰り返し行う高度な分析機能です。工数のかかるレポート作成やマニュアル作成を短時間でAIが作成します。
RAG Chatが「知識から回答するAI」に対し、Deep Researchは 「業務を実行するAI」です。ユーザーから与えられたタスクに対して、いくつもの文書やデータソースを横断し、AIが自律的に調査を計画→実行→考察を繰り返し行う高度な分析機能です。工数のかかるレポート作成やマニュアル作成を短時間でAIが作成します。
Whereas RAG Chat is an “AI that answers from knowledge,” Deep Research is an “AI that executes work.” For tasks given by the user, it is an advanced analysis function in which the AI autonomously plans, executes, and reflects on its research across many documents and data sources. It enables the AI to create labor-intensive reports and manuals in a short time.
Batch Analysis
Batch Analysis は、大量の質問やチェック項目のリストに対して、AIが一件ずつ順番に処理し、回答や判定を生成する機能です。一問一答形式のチャットでは時間がかかる作業でも、リストを投入するだけで AIが各項目を個別に解析し、同じ精度・同じ基準で回答や評価を返します。
Batch Analysis は、大量の質問やチェック項目のリストに対して、AIが一件ずつ順番に処理し、回答や判定を生成する機能です。一問一答形式のチャットでは時間がかかる作業でも、リストを投入するだけで AIが各項目を個別に解析し、同じ精度・同じ基準で回答や評価を返します。
Batch Analysis lets AI process lists of questions or check items one by one, generating an answer and judgment for each. Instead of time‑consuming one‑by‑one chats, you submit a list and the AI analyzes items individually, returning answers and evaluations with consistent accuracy and standards.
RAG Chat
Buried in data but missing the answers you need? AskDona's RAG-powered chat unifies your organization's diverse data into a single hub, classifies it, and organizes it for optimal retrieval; for each question, AskDona probes from multiple angles and follows a reason → retrieve → synthesize loop to deliver clear reliable, answers.
RAG Chat
Buried in data but missing the answers you need? AskDona's RAG-powered chat unifies your organization's diverse data into a single hub, classifies it, and organizes it for optimal retrieval; for each question, AskDona probes from multiple angles and follows a reason → retrieve → synthesize loop to deliver clear reliable, answers.
DeepResearch
RAG Chat is an AI that answers from knowledge, while Deep Research is an AI that performs business tasks. Given a user’s task, it performs advanced analysis by autonomously cycling through planning, executing and reflecting across many documents and data sources. It quickly produces labor‑intensive deliverables—such as reports and manuals—that would otherwise take a lot of time and effort, in a short amount of time.
DeepResearch
RAG Chat is an AI that answers from knowledge, while Deep Research is an AI that performs business tasks. Given a user’s task, it performs advanced analysis by autonomously cycling through planning, executing and reflecting across many documents and data sources. It quickly produces labor‑intensive deliverables—such as reports and manuals—that would otherwise take a lot of time and effort, in a short amount of time.
Batch Analysis
Batch Analysis is a feature where the AI processes large lists of questions or checklist items one by one, generating answers or decisions. What would be slow in one‑by‑one chat becomes simple: drop in a list, and the AI analyzes each item individually, returning responses and evaluations with consistent accuracy and uniform criteria.
Batch Analysis
Batch Analysis is a feature where the AI processes large lists of questions or checklist items one by one, generating answers or decisions. What would be slow in one‑by‑one chat becomes simple: drop in a list, and the AI analyzes each item individually, returning responses and evaluations with consistent accuracy and uniform criteria.
RAG Chat
Even with mountains of documents, you still can’t find the answer you need. RAG Chat solves this by bringing many kinds of internal data together in one place, classifying it, and organizing it in an optimal way. Users can access the organization’s knowledge at any time and get the information they need.
RAG Chat
Even with mountains of documents, you still can’t find the answer you need. RAG Chat solves this by bringing many kinds of internal data together in one place, classifying it, and organizing it in an optimal way. Users can access the organization’s knowledge at any time and get the information they need.
Deep Research
Whereas RAG Chat is an “AI that answers from knowledge,” Deep Research is an “AI that executes work.” For tasks given by the user, it is an advanced analysis function in which the AI autonomously plans, executes, and reflects on its research across many documents and data sources. It enables the AI to create labor-intensive reports and manuals in a short time.
Deep Research
Whereas RAG Chat is an “AI that answers from knowledge,” Deep Research is an “AI that executes work.” For tasks given by the user, it is an advanced analysis function in which the AI autonomously plans, executes, and reflects on its research across many documents and data sources. It enables the AI to create labor-intensive reports and manuals in a short time.
Batch Analysis
Batch Analysis lets AI process lists of questions or check items one by one, generating an answer and judgment for each. Instead of time‑consuming one‑by‑one chats, you submit a list and the AI analyzes items individually, returning answers and evaluations with consistent accuracy and standards.
Batch Analysis
Batch Analysis lets AI process lists of questions or check items one by one, generating an answer and judgment for each. Instead of time‑consuming one‑by‑one chats, you submit a list and the AI analyzes items individually, returning answers and evaluations with consistent accuracy and standards.
Next-Generation AI for Complex Queries
At AskDona, we define “complex questions” as those that cannot be answered from a single source, but require information spread across multiple documents to be explored and integrated from multiple perspectives, and then logically reconstructed.
AskDona’s “dona-rag-2.0” is built on a proprietary architecture that goes beyond traditional RAG (Retrieval-Augmented Generation) based on single search results. This allows the system to deliver highly comprehensive answers even to complex questions posed by organizational experts.
Next-Generation AI for Complex Queries
At AskDona, we define “complex questions” as those that cannot be answered from a single source, but require information spread across multiple documents to be explored and integrated from multiple perspectives, and then logically reconstructed.
AskDona’s “dona-rag-2.0” is built on a proprietary architecture that goes beyond traditional RAG (Retrieval-Augmented Generation) based on single search results. This allows the system to deliver highly comprehensive answers even to complex questions posed by organizational experts.
Next-Generation AI for Complex Queries
At AskDona, we define “complex questions” as those that cannot be answered from a single source, but require information spread across multiple documents to be explored and integrated from multiple perspectives, and then logically reconstructed.
AskDona’s “dona-rag-2.0” is built on a proprietary architecture that goes beyond traditional RAG (Retrieval-Augmented Generation) based on single search results. This allows the system to deliver highly comprehensive answers even to complex questions posed by organizational experts.
Next-Generation AI for Complex Queries
At AskDona, we define “complex questions” as those that cannot be answered from a single source, but require information spread across multiple documents to be explored and integrated from multiple perspectives, and then logically reconstructed.
AskDona’s “dona-rag-2.0” is built on a proprietary architecture that goes beyond traditional RAG (Retrieval-Augmented Generation) based on single search results. This allows the system to deliver highly comprehensive answers even to complex questions posed by organizational experts.
SEARCH. THINK. ANSWER.
SEARCH. THINK. ANSWER.
SEARCH. THINK. ANSWER.
Providing evidence-based, highly accurate answers in real time for advanced information.
Providing evidence-based, highly accurate answers in real time for advanced information.
Providing evidence-based, highly accurate answers in real time for advanced information.
Enterprise-Grade Document Preprocessing
Enterprise-Grade Document Preprocessing
Image and Diagram Analysis
Charts & graphs, flowcharts, diagrams and screenshots.
Image and Diagram Analysis
Charts & graphs, flowcharts, diagrams and screenshots.
Image and Diagram Analysis
Charts & graphs, flowcharts, diagrams and screenshots.
Image and Diagram Analysis
Charts & graphs, flowcharts, diagrams and screenshots.
Table and Formula Extraction
Table and Formula Extraction
Complex tables, mathematical formulas, financial statements and scientific notation.
Complex tables, mathematical formulas, financial statements and scientific notation.
Semantic Chunking
Semantic Chunking
Context-driven semantic splitting of text into chunks without any information loss.
Context-driven semantic splitting of text into chunks without any information loss.
Image and Diagram Analysis
Image and Diagram Analysis
Charts & graphs, flowcharts, diagrams and screenshots.
Charts & graphs, flowcharts, diagrams and screenshots.
Metadata Preservation
Metadata Preservation
Original filename & format, document properties e.g. upload date, user, page no.
Original filename & format, document properties e.g. upload date, user, page no.
What file formats does AskDona support? Can it also read scanned documents depending on their quality?
Original filename & format, document properties e.g. upload date, user, page no.
Table and Formula Extraction
Complex tables, mathematical formulas, financial statements and scientific notation.
Semantic Chunking
Context-driven semantic splitting of text into chunks without any information loss.
Image and Diagram Analysis
Charts & graphs, flowcharts, diagrams and screenshots.
Metadata Preservation
Original filenames & format, document properties e.g. upload date, user, page no.
Image and Diagram Analysis
Charts & graphs, flowcharts, diagrams and screenshots.
Image and Diagram Analysis
Charts & graphs, flowcharts, diagrams and screenshots.
Turning 1 question into multiple subqueries:
• AI agents break down complex or ambiguous questions into clearer components and convert them into sub-queries.
• More accurate and complete information retrieval.
• Reduces hallucinations by grounding each part of the query.
01
Subqueries
Combination of vectors and keywords:
• Pure vector search may miss rare words, dates, or brand names.
• Pure keyword search misses meaning and context.
• Hybrid search gets the strengths of both.
02
Hybrid Search
Every response from RAG chat incudes:
•Inline citations such as numbers [1],[2],etc. linking to sources.
•Source cards, expandable previews and direct links.
•This prevents hallucination, builds trust, enables fact-checking and supports compliance.
03
Citation Mapping
Turning 1 question into multiple subqueries:
• AI agents break down complex or ambiguous questions into clearer components and convert them into sub-queries.
• More accurate and complete information retrieval.
• Reduces hallucinations by grounding each part of the query.
01
Subqueries
Combination of vectors and keywords:
• Pure vector search may miss rare words, dates, or brand names.
• Pure keyword search misses meaning and context.
• Hybrid search gets the strengths of both.
02
Hybrid Search
Every response from RAG chat incudes:
•Inline citations such as numbers [1],[2],etc. linking to sources.
•Source cards, expandable previews and direct links.
•This prevents hallucination, builds trust, enables fact-checking and supports compliance.
03
Citation Mapping
What file formats does AskDona support? Can it also read scanned documents depending on their quality?
Turning 1 question into multiple subqueries:
• AI agents break down complex or ambiguous questions into clearer components and convert them into sub-queries.
• More accurate and complete information retrieval.
• Reduces hallucinations by grounding each part of the query.
01
Subqueries
Combination of vectors and keywords:
• Pure vector search may miss rare words, dates, or brand names.
• Pure keyword search misses meaning and context.
• Hybrid search gets the strengths of both.
02
Hybrid Search
Every response from RAG chat incudes:
•Inline citations such as numbers [1],[2],etc. linking to sources.
•Source cards, expandable previews and direct links.
•This prevents hallucination, builds trust, enables fact-checking and supports compliance.
03
Citation Mapping
Turning 1 question into multiple subqueries:
• AI agents break down complex or ambiguous questions into clearer components and convert them into sub-queries.
• More accurate and complete information retrieval.
• Reduces hallucinations by grounding each part of the query.
01
Subqueries
Combination of vectors and keywords:
• Pure vector search may miss rare words, dates, or brand names.
• Pure keyword search misses meaning and context.
• Hybrid search gets the strengths of both.
02
Hybrid Search
Every response from RAG chat incudes:
•Inline citations such as numbers [1],[2],etc. linking to sources.
•Source cards, expandable previews and direct links.
•This prevents hallucination, builds trust, enables fact-checking and supports compliance.
03
Citation Mapping
Turning 1 question into multiple subqueries:
• AI agents break down complex or ambiguous questions into clearer components and convert them into sub-queries.
• More accurate and complete information retrieval.
• Reduces hallucinations by grounding each part of the query.
01
Subqueries
Combination of vectors and keywords:
• Pure vector search may miss rare words, dates, or brand names.
• Pure keyword search misses meaning and context.
• Hybrid search gets the strengths of both.
02
Hybrid Search
Every response from RAG chat incudes:
•Inline citations such as numbers [1],[2],etc. linking to sources.
•Source cards, expandable previews and direct links.
•This prevents hallucination, builds trust, enables fact-checking and supports compliance.
03
Citation Mapping
Yes. AskDona accepts Microsoft Excel, Word, and PowerPoint files, as well as image files (JPEG, PNG, GIF, etc.) and text files (TXT, CSV, JSON, etc.). It can also handle scanned data, depending on the resolution.
Advanced Real-Time Processing
Advanced Real-Time Processing
Advanced Real-Time Processing
Yes. AskDona accepts Microsoft Excel, Word, and PowerPoint files, as well as image files (JPEG, PNG, GIF, etc.) and text files (TXT, CSV, JSON, etc.). It can also handle scanned data, depending on the resolution.
High accuracy from Day 1 is the new standard.
High accuracy from Day 1 is the new standard.
AskDona’s “dona-rag-2.0” delivers high-accuracy answers from the initial phase without requiring tuning or data preprocessing. Joint verification with RIKEN R-CCS, the organization operating the supercomputer “Fugaku,” has demonstrated that it achieves accuracy more than 20 points higher than competing products.
Instead of spending time on accuracy improvements, the person in charge of implementation can focus fully on creating value by leveraging AskDona as organizational knowledge. By resolving technical concerns early, the verification phase can be cleared quickly, enabling a smooth transition to full-scale operation without causing opportunity loss.
AskDona’s “dona-rag-2.0” delivers high-accuracy answers from the initial phase without requiring tuning or data preprocessing. Joint verification with RIKEN R-CCS, the organization operating the supercomputer “Fugaku,” has demonstrated that it achieves accuracy more than 20 points higher than competing products.
Instead of spending time on accuracy improvements, the person in charge of implementation can focus fully on creating value by leveraging AskDona as organizational knowledge. By resolving technical concerns early, the verification phase can be cleared quickly, enabling a smooth transition to full-scale operation without causing opportunity loss.
High accuracy from Day 1 is the new standard.
AskDona’s “dona-rag-2.0” delivers high-accuracy answers from the initial phase without requiring tuning or data preprocessing. Joint verification with RIKEN R-CCS, the organization operating the supercomputer “Fugaku,” has demonstrated that it achieves accuracy more than 20 points higher than competing products.
Instead of spending time on accuracy improvements, the person in charge of implementation can focus fully on creating value by leveraging AskDona as organizational knowledge. By resolving technical concerns early, the verification phase can be cleared quickly, enabling a smooth transition to full-scale operation without causing opportunity loss.
Yes. AskDona accepts Microsoft Excel, Word, and PowerPoint files, as well as image files (JPEG, PNG, GIF, etc.) and text files (TXT, CSV, JSON, etc.). It can also handle scanned data, depending on the resolution.
AskDona is an extremely useful tool for providing accurate and necessary information to Fugaku users. We look forward to its further development in the future.
Fumiyoshi Shoji
Director, Operations Technology Division, RIKEN Center for Computational Science
I’m building a new website and it’s absolutely ridiculous how valuable your content has been.
Michael Riddering
Ask Dona has been helpful
James Traf
AskDona is an extremely useful tool for providing accurate and necessary information to Fugaku users. We look forward to its further development in the future.
Fumiyoshi Shoji
Director, Operations Technology Division, RIKEN Center for Computational Science
I’m building a new website and it’s absolutely ridiculous how valuable your content has been.
Michael Riddering
Ask Dona has been helpful
James Traf
AskDona is an extremely useful tool for providing accurate and necessary information to Fugaku users. We look forward to its further development in the future.
Fumiyoshi Shoji
Director, Operations Technology Division, RIKEN Center for Computational Science
I’m building a new website and it’s absolutely ridiculous how valuable your content has been.
Michael Riddering
Ask Dona has been helpful
James Traf
AskDona is an extremely useful tool for providing accurate and necessary information to Fugaku users. We look forward to its further development in the future.
Fumiyoshi Shoji
Director, Operations Technology Division, RIKEN Center for Computational Science
I’m building a new website and it’s absolutely ridiculous how valuable your content has been.
Michael Riddering
Ask Dona has been helpful
James Traf
AskDona is an extremely useful tool for providing accurate and necessary information to Fugaku users. We look forward to its further development in the future.
Fumiyoshi Shoji
Director, Operations Technology Division, RIKEN Center for Computational Science
I’m building a new website and it’s absolutely ridiculous how valuable your content has been.
Michael Riddering
Ask Dona has been helpful
James Traf
AskDona Features
AskDona Features
AskDona Features
Stay up to date with the latest news on AskDona, insights on generative AI, RAG, security, ROI, and cutting-edge technological trends.
Stay up to date with the latest news on AskDona, insights on generative AI, RAG, security, ROI, and cutting-edge technological trends.
Stay up to date with the latest news on AskDona, insights on generative AI, RAG, security, ROI, and cutting-edge technological trends.
August 1, 2025
What is "RAG (Search Augmentation)"? An approach to improving answer accuracy
What is "RAG (Search Augmentation)"? An approach to improving answer accuracy
Read more
Read more
August 2, 2025
Growth speed and technological innovation of the generative AI market in 2025. Security measures and ROI concerns for companies.
Growth speed and technological innovation of the generative AI market in 2025. Security measures and ROI concerns for companies.
Read more
Read more
Frequently asked questions
Frequently asked questions
Frequently asked questions
Will the data I upload to the RAG database be stored on a domestic server?
Yes, files uploaded to the RAG database are managed on servers within Japan.
Will the data I upload to the RAG database be stored on a domestic server?
Yes, files uploaded to the RAG database are managed on servers within Japan.
Will the data I upload to the RAG database be stored on a domestic server?
Yes, files uploaded to the RAG database are managed on servers within Japan.
Will the data I upload to the RAG database be stored on a domestic server?
Yes, files uploaded to the RAG database are managed on servers within Japan.
Is the AskDona usage environment separated from other companies?
Is the AskDona usage environment separated from other companies?
Is the AskDona usage environment separated from other companies?
Is the AskDona usage environment separated from other companies?
Can I use AskDona in my own cloud environment?
Can I use AskDona in my own cloud environment?
Can I use AskDona in my own cloud environment?
Can I use AskDona in my own cloud environment?
What are the file formats and capacity limits that can be stored in RAG?
What are the file formats and capacity limits that can be stored in RAG?
What are the file formats and capacity limits that can be stored in RAG?
What are the file formats and capacity limits that can be stored in RAG?
Will internal data be used to train ChatGPT?
Will internal data be used to train ChatGPT?
Will internal data be used to train ChatGPT?
Will internal data be used to train ChatGPT?
Start with a free consultation.
AskDona is dedicated to Redefining Roles. We empower your organization by assigning all AI-capable tasks to the technology, allowing you to focus on critical human functions and innovation.
Start with a free consultation.
AskDona is dedicated to Redefining Roles. We empower your organization by assigning all AI-capable tasks to the technology, allowing you to focus on critical human functions and innovation.
Start with a free consultation.
AskDona is dedicated to Redefining Roles. We empower your organization by assigning all AI-capable tasks to the technology, allowing you to focus on critical human functions and innovation.
Start with a free consultation.
AskDona is dedicated to Redefining Roles. We empower your organization by assigning all AI-capable tasks to the technology, allowing you to focus on critical human functions and innovation.