Get the newsletter
Microsoft · AI / ML · Filed Jun 20, 2025 · USPTO published Nov 19, 2026 · This week's release

Microsoft Patents a Filter That Decides Which Files and Chats Its AI Reads

Microsoft's patent describes a research-assistant step for work AI. It sorts your meetings, chats and files, tosses duplicates and contradictions, and remembers its work for next time.

Software that sifts through files and chats to pull out the most relevant details before sending them to an AI assistant. Drawing from patent filing US 2026/0368186 A1.
Software that sifts through files and chats to pull out the most relevant details before sending them to an AI assistant.All 20 drawings ↓
Publication number
US 2026/0368186 A1(US20260368186A1)
Applicant
MICROSOFT TECHNOLOGY LICENSING, LLC
Filed · Published
Jun 20, 2025 · Nov 19, 2026
Inventors
Sakshi MUNJAL, . CHAITANYA, Priyank RASTOGI, Sachin SWAMI, Rohan GYANI, Sheetal S SETHI, Sandeep Kumar SRIVASTAVA
Status when we published
Approved; patent expected soon
Document
20 claims
US classification
707/723
Examiner
LEWIS, CHERYL RENEA (Art Unit 2166)

What Microsoft's AI information sorter actually does

Imagine you ask a work AI assistant to draft a project update. Behind the scenes it could pull from last month's meeting recordings, hundreds of chat messages, shared files and web pages. Most of that is noise, and some of it contradicts itself.

Microsoft's patent describes a sorting system that sits between those sources and the AI doing the work. It breaks long material into smaller pieces, labels each piece with a topic and where it came from, and scores how useful it is for the task. It throws out duplicates and anything that doesn't connect to other relevant material. What's left gets stitched into one clean briefing.

The system can also ask its own questions to fill gaps in what it knows. It remembers its work for a while, so the next task doesn't start from scratch.

From the filing · CLAIM 1
… ranking and filtering out irrelevant contexts; adding a re-ranking and filtering phase; and synthesizing remaining contexts into a usable format.

Translation: The system sorts through data multiple times to drop useless information before combining what is left.

How chunks get scored, re-ranked and cached

The patent covers a pipeline for the grounding step in Microsoft's Multi-Agent Runtime Service, which is used to build and run teams of AI agents. Grounding means handing an AI model the background information it needs, such as meeting recordings, chats, files and planner data. The filing says pulling from so many places adds delay and raises the risk of using wrong or irrelevant data.

The pipeline works in stages:

  • Gather: a meta grounding agent pulls possibly relevant material from every source, tuned for speed. Each item carries notes on what it is, where it came from and the context it was shared in.
  • Split: a preprocessor agent (an AI model assigned to this job) cuts long text into chunks of whatever size it judges best. It tags each chunk with topics and a source link for citations.
  • Rank and filter: a reasoning agent groups chunks by topic, scores them on task relevance and source credibility, and drops duplicates and conflicts.
  • Re-rank: a second pass updates the scores using the relationships between chunks. It drops low scorers and anything that clashes with high scorers.
  • Synthesize: what remains is joined into one grounding context for the agents doing the task.

The filing also describes an index, so each execution agent reads only the parts it needs on its turn and can follow a link back to the source. Selected results can be saved for a chosen time window, so repeat tasks run faster. A questioning agent can generate questions to fill in missing information, and a knowledge graph can map how chunks relate.

From the filing · THE ABSTRACT
Depending on the signals from initial phases, the processing pipeline is expanded to perform preprocessing over extracted grounding context and splitting data into chunks with context as needed.

Translation: The software breaks files into smaller pieces and processes them dynamically based on early results.

Why workplace AI answers depend on this step

If you use an AI assistant at work, the quality of its answers depends on what it reads first. Feed it too much and it slows down or loses detail. Feed it the wrong paragraph and it answers wrongly with total confidence. Sorting, double-checking and saving earlier work aims to make answers faster and less likely to rest on stale or irrelevant material.

The patent application has been approved by the patent office. Once Microsoft pays the final fee, it should be issued as a granted patent. The filing ties the idea to Microsoft's Multi-Agent Runtime Service, so it reads as groundwork for AI agents that handle tasks across your meetings, chats and files.

One more thing to know: the patent office has already approved this application. Once Microsoft pays the final fee, it should be issued as a granted patent, and the final claims may read a little differently from the version described here.

This is the 22nd Microsoft filing we've tracked since June on our assistants that remember you watchlist, following one on ignored email topics and one on coding context.

Source: US 2026/0368186 A1 at the USPTO (PDF). Full patent text and figures are in the official publication.

Get our take in your Top Stories

Liked this breakdown? Add Patentlyze as a preferred source on Google, and our plain-English take shows up more often in your Top Stories the next time Microsoft patent news breaks.

There are more where this came from

We read every patent application Big Tech publishes and send you the ones worth knowing. Plain English, free, every week.

The drawings · 20 sheets

20 drawing sheets from US 2026/0368186 A1 · click any drawing to enlarge

Turning raw files into clear facts for AI, drawing sheet 1 from US 2026/0368186 A1
Sheet 1 Turning raw files into clear facts for AI
Automated agents question and verify facts found in your files, drawing sheet 2 from US 2026/0368186 A1
Sheet 2 Automated agents question and verify facts found in your files
Connecting verified facts to apps to finish a task, drawing sheet 3 from US 2026/0368186 A1
Sheet 3 Connecting verified facts to apps to finish a task
Breaking long documents into smaller ranked snippets for the AI, drawing sheet 4 from US 2026/0368186 A1
Sheet 4 Breaking long documents into smaller ranked snippets for the AI
Picking specialized helper programs to search emails and transcripts, drawing sheet 5 from US 2026/0368186 A1
Sheet 5 Picking specialized helper programs to search emails and transcripts
Fetching actual stored files and notes through cloud connectors, drawing sheet 6 from US 2026/0368186 A1
Sheet 6 Fetching actual stored files and notes through cloud connectors
Sample rules defining which meeting notes and files to pull, drawing sheet 7 from US 2026/0368186 A1
Sheet 7 Sample rules defining which meeting notes and files to pull
Asking automated questions to fill in missing details, drawing sheet 8 from US 2026/0368186 A1
Sheet 8 Asking automated questions to fill in missing details
Show the other 12 sheets
How AI assistants coordinate to collect background facts, drawing sheet 9 from US 2026/0368186 A1
Sheet 9 How AI assistants coordinate to collect background facts
Slicing text into pieces and selecting the most relevant parts, drawing sheet 10 from US 2026/0368186 A1
Sheet 10 Slicing text into pieces and selecting the most relevant parts
Resolving contradictory notes before assembling the final summary, drawing sheet 11 from US 2026/0368186 A1
Sheet 11 Resolving contradictory notes before assembling the final summary
Passing unprocessed source data directly into the reasoning stage, drawing sheet 12 from US 2026/0368186 A1
Sheet 12 Passing unprocessed source data directly into the reasoning stage
Storing searchable notes in memory for AI tools, drawing sheet 13 from US 2026/0368186 A1
Sheet 13 Storing searchable notes in memory for AI tools
Pulling facts from files while removing irrelevant details, drawing sheet 14 from US 2026/0368186 A1
Sheet 14 Pulling facts from files while removing irrelevant details
How the system breaks down documents and ranks the snippets, drawing sheet 15 from US 2026/0368186 A1
Sheet 15 How the system breaks down documents and ranks the snippets
Indexing text so AI helpers can take turns reading, drawing sheet 16 from US 2026/0368186 A1
Sheet 16 Indexing text so AI helpers can take turns reading
Breaking up and ranking background files for an AI, drawing sheet 17 from US 2026/0368186 A1
Sheet 17 Breaking up and ranking background files for an AI
Hardware parts that process data and run the software, drawing sheet 18 from US 2026/0368186 A1
Sheet 18 Hardware parts that process data and run the software
How an app sends work to cloud servers and AI, drawing sheet 19 from US 2026/0368186 A1
Sheet 19 How an app sends work to cloud servers and AI
Connecting phones, tablets, and computers to an AI host system, drawing sheet 20 from US 2026/0368186 A1
Sheet 20 Connecting phones, tablets, and computers to an AI host system
Patent filing page

More in AI assistant & agent

All 436 AI assistant & agent patents →

Watchlists this filing belongs to

All watchlists →
Reader comments

No comments yet

Start the discussion

Real name or a handle, either is fine. Comments are read by a person before they appear, so allow a little time. Keep it about the filing.