Meeting AI Assistant "Kaigi Copilot" Offering Real-time AI Support and Local-first Privacy Design, Launched by Incerto
Incerto LLC has launched "Kaigi Copilot," a meeting AI assistant that integrates real-time AI support during meetings with a local-first privacy design, ensuring data never leaves the company's internal environment. This solution allows confidential meetings to leverage AI without compromising security.
📋 Article Processing Timeline
- 📰 Published: May 9, 2026 at 02:19
- 🔍 Collected: May 8, 2026 at 17:32
- 🤖 AI Analyzed: May 8, 2026 at 19:49 (2h 17m after Collected)
Incerto LLC (Headquarters: Arakawa-ku, Tokyo, Representative: Aoto Sato, https://www.incerto.tech/) today announced the launch of "Kaigi Copilot," a meeting AI assistant designed for highly confidential meetings. Product details are available on the product page (https://www.incerto.tech/products/meeting-copilot).
"Kaigi Copilot" is a meeting AI assistant that seamlessly integrates two core features: real-time AI support, which analyzes speech and screen sharing during meetings to provide suggested answers, discussion points, and related information in real time; and a local-first privacy design, ensuring that recording, transcription, meeting minutes, and AI analysis are all completed within the company's internal environment without being sent to external clouds. These features are provided together in a single application.
The application operates independently of meeting tools (such as Zoom, Google Meet, and Microsoft Teams), running persistently on the user's device. This allows for real-time AI support and automated meeting minutes even in meetings where external cloud-based AI services are difficult to use, such as business negotiations, management meetings, HR interviews, legal consultations, and R&D reviews. The underlying AI analysis can be configured in a hybrid manner, combining local LLMs, on-premise AI, and enterprise cloud AI, chosen based on the confidentiality level of the meeting and the operational requirements of the IT department.
■ Background of Provision
AI-powered business support has primarily expanded to tasks "after" and "before" meetings, such as creating meeting minutes, organizing reports, and analyzing data. Recently, services have begun to emerge that support the actual speech and judgment "during" meetings.
However, the very meetings that most need AI support are often those where external cloud-based AI services are difficult to use. This is because meetings with significant value from AI's immediate response, reference, and organization capabilities—such as management decisions, performance evaluations, contract negotiations, intellectual property, sensitive customer information, medical/legal consultations, and R&D inventions—often coincide with meetings that have restrictions on sending data to external clouds or processing it on third-party platforms.
"Kaigi Copilot" was designed with this situation in mind. Its design starts from the premise of enabling real-time AI support during meetings and automated meeting minutes, all while keeping meeting data within the company and never sending it to external clouds.
■ Challenge: Opacification of Organizational Business Practices
In meeting environments, the responsiveness of participants increasingly determines outcomes across all industries. Immediate answers to unexpected questions, instant reference to past meeting minutes or recent figures, and retrieval of internal knowledge directly impact the success of business negotiations, the speed of management decisions, and the quality of decision-making in specialized meetings.
These are also tasks that AI excels at. The process of instantly retrieving information most relevant to the current discussion context from a vast repository of past meeting minutes, internal documents, and FAQs is limited for humans alone. AI intervention can significantly elevate the quality of meetings.
However, the very meetings where AI support offers the most value are often highly confidential, preventing the use of meeting AI assistants that rely on external cloud transmission. Sending audio from a management meeting to an external AI is not feasible, explaining to a client during a business negotiation that "data will be sent to an external cloud" is difficult, and in HR interviews, legal consultations, and specialized meetings, sensitive information cannot be passed through external platforms. These restrictions have kept the meetings that could benefit most from AI support the furthest away from it.
Many companies currently ensure meeting quality through a combination of manager attendance, manual meeting minutes, and human support. However, the quality of on-site response depends on the individual experience of the attendees, making it difficult to ensure reproducibility. It is not realistic for sales managers to attend all business negotiations, nor for executives to retrieve all internal knowledge in every specialized meeting.
In other words, as long as organizations rely on external cloud-based AI for improving meeting quality, these solutions cannot reach confidential meetings. This single point represents the structural challenge that this release addresses.
■ Solution: "Kaigi Copilot"
"Kaigi Copilot" addresses this structural challenge by offering two core functions of a meeting AI assistant—real-time AI support during meetings and local-first privacy design—integrated into a single application.
The meeting-specific application runs on the user's PC, operating independently and persistently alongside meeting tools like Zoom, Google Meet, and Microsoft Teams. During meetings, it analyzes speech and screen sharing to provide real-time support, and after meetings, it stores recordings, transcriptions, and meeting minutes within the same application. The design offers the option to ensure that recording data, transcriptions, meeting minutes, and AI analysis pathways are never sent to external clouds.
"Kaigi Copilot" is a meeting AI assistant that seamlessly integrates two core features: real-time AI support, which analyzes speech and screen sharing during meetings to provide suggested answers, discussion points, and related information in real time; and a local-first privacy design, ensuring that recording, transcription, meeting minutes, and AI analysis are all completed within the company's internal environment without being sent to external clouds. These features are provided together in a single application.
The application operates independently of meeting tools (such as Zoom, Google Meet, and Microsoft Teams), running persistently on the user's device. This allows for real-time AI support and automated meeting minutes even in meetings where external cloud-based AI services are difficult to use, such as business negotiations, management meetings, HR interviews, legal consultations, and R&D reviews. The underlying AI analysis can be configured in a hybrid manner, combining local LLMs, on-premise AI, and enterprise cloud AI, chosen based on the confidentiality level of the meeting and the operational requirements of the IT department.
■ Background of Provision
AI-powered business support has primarily expanded to tasks "after" and "before" meetings, such as creating meeting minutes, organizing reports, and analyzing data. Recently, services have begun to emerge that support the actual speech and judgment "during" meetings.
However, the very meetings that most need AI support are often those where external cloud-based AI services are difficult to use. This is because meetings with significant value from AI's immediate response, reference, and organization capabilities—such as management decisions, performance evaluations, contract negotiations, intellectual property, sensitive customer information, medical/legal consultations, and R&D inventions—often coincide with meetings that have restrictions on sending data to external clouds or processing it on third-party platforms.
"Kaigi Copilot" was designed with this situation in mind. Its design starts from the premise of enabling real-time AI support during meetings and automated meeting minutes, all while keeping meeting data within the company and never sending it to external clouds.
■ Challenge: Opacification of Organizational Business Practices
In meeting environments, the responsiveness of participants increasingly determines outcomes across all industries. Immediate answers to unexpected questions, instant reference to past meeting minutes or recent figures, and retrieval of internal knowledge directly impact the success of business negotiations, the speed of management decisions, and the quality of decision-making in specialized meetings.
These are also tasks that AI excels at. The process of instantly retrieving information most relevant to the current discussion context from a vast repository of past meeting minutes, internal documents, and FAQs is limited for humans alone. AI intervention can significantly elevate the quality of meetings.
However, the very meetings where AI support offers the most value are often highly confidential, preventing the use of meeting AI assistants that rely on external cloud transmission. Sending audio from a management meeting to an external AI is not feasible, explaining to a client during a business negotiation that "data will be sent to an external cloud" is difficult, and in HR interviews, legal consultations, and specialized meetings, sensitive information cannot be passed through external platforms. These restrictions have kept the meetings that could benefit most from AI support the furthest away from it.
Many companies currently ensure meeting quality through a combination of manager attendance, manual meeting minutes, and human support. However, the quality of on-site response depends on the individual experience of the attendees, making it difficult to ensure reproducibility. It is not realistic for sales managers to attend all business negotiations, nor for executives to retrieve all internal knowledge in every specialized meeting.
In other words, as long as organizations rely on external cloud-based AI for improving meeting quality, these solutions cannot reach confidential meetings. This single point represents the structural challenge that this release addresses.
■ Solution: "Kaigi Copilot"
"Kaigi Copilot" addresses this structural challenge by offering two core functions of a meeting AI assistant—real-time AI support during meetings and local-first privacy design—integrated into a single application.
The meeting-specific application runs on the user's PC, operating independently and persistently alongside meeting tools like Zoom, Google Meet, and Microsoft Teams. During meetings, it analyzes speech and screen sharing to provide real-time support, and after meetings, it stores recordings, transcriptions, and meeting minutes within the same application. The design offers the option to ensure that recording data, transcriptions, meeting minutes, and AI analysis pathways are never sent to external clouds.