Muse Draws Attention as Questions Resurface About Its Similarities to OpenClaw

What Muse Is and What We Know About Its Launch

Meta introduced Muse on September 8, 2026, describing it as a personal artificial intelligence agent. The company presented it as a product for a general audience and said it had been built from scratch. Associated Press coverage also reported Meta’s emphasis on safety and privacy. These descriptions help establish what the product is and how the company positions it, but they do not, by themselves, measure its performance or independently verify its capabilities.

The distinction between an agent and an assistant that only answers questions matters when considering the interest Muse has attracted: an agent is intended to carry out tasks, not just generate text. However, the sources available here do not provide enough detail to establish which specific tasks were available to all users, under what conditions the agent could perform them, or what confirmations it required. It would therefore be unwise to present individual use cases as general capabilities unless they are documented in the verified sources.

Attention around the launch increased after news outlets reported a high level of early use during the first week. That figure needs careful interpretation: where it came from, how it was defined, and what period it covers all matter as much as the number itself.

The Adoption Figure: People Who Tried It Are Not the Same as Downloads

The Decoder reported that Muse reached more than 500,000 people during its first week. According to that coverage, the figure came from internal information reported by The Information. This means it is not a primary measurement published by an independent organization in the materials available here, nor is it a figure that can be directly attributed to an app store.

It is also important not to treat “people who tried it,” “users,” and “downloads” as interchangeable measures. Someone can download an app without opening it; a download can be repeated across different devices; and “user” may mean someone who launched the app, created an account, or took some other action. The available article does not explain exactly how the figure of more than 500,000 was defined or how it was calculated. The figure therefore offers a reported signal of early interest, but it does not show how many people used Muse regularly.

In a later article, TechCrunch compared Muse’s early trajectory with ChatGPT’s using estimates from Sensor Tower and Apptopia. That coverage discusses downloads and compares product launches; it does not confirm the number of people who tried Muse. These are different metrics and should be kept separate. Nor does a high position in an app-store ranking, on its own, tell us about retention, satisfaction, or sustained use.

The practical conclusion should be limited: reports point to a substantial initial response, but the sources provided do not offer a detailed public measurement that establishes the precise reach, recurring activity, or make-up of Muse’s audience.

What Has Been Reported About OpenClaw

The controversy centers on reported similarities between elements in Muse’s workspace and files associated with OpenClaw. TechCrunch reported that some people had pointed to matching filenames and similar content, highlighting SOUL.md as an example. The Decoder also described filenames and file contents as nearly identical. These are journalistic reports about a comparison; the sources provided do not include the relevant Muse files or a reproducible technical analysis that would allow readers to verify each match for themselves.

OpenClaw’s public documentation does make it possible to confirm that SOUL.md is among the concepts documented by the project. The official repository can also be inspected for OpenClaw’s files and structure. But knowing what is in the OpenClaw materials is not enough to confirm that a Muse file contains the same text, or to determine how it got there. That would require the relevant versions from both sides, with their dates and provenance identified, as well as a comparison method that others could reproduce.

The statement attributed to Meta by TechCrunch credits Nat Friedman, Meta Superintelligence Labs’ head of product, with making two points that should be read together: Muse was built from scratch, and, as a product, it is “heavily inspired” by OpenClaw. TechCrunch reported that Meta referred the outlet to that statement. This is an acknowledgment of inspiration, not an admission that code, protected content, or the project as a whole was copied.

The distinction matters. Declared inspiration describes an influence the company has acknowledged. An observed similarity is a claim that particular features appear to match. Copying, by contrast, would require identifying what was reproduced and the evidence for that claim; a conclusion about infringement would also require a legal analysis of the specific elements involved. The available sources do not establish such a conclusion.

What Is Confirmed, What Is Attributed, and What Remains Unresolved

Based on the available information, three levels can be distinguished. First, facts about the announcement: Meta introduced Muse on September 8, 2026, and described it as a personal agent built from scratch. Second, attributed statements and reports: according to TechCrunch, Meta acknowledged strong inspiration from OpenClaw; several outlets reported matches in filenames and file contents; and The Decoder attributed the figure of more than 500,000 people in the first week to internal information. Third, unresolved questions: the exact technical comparison, the definition of the user metric, and whether any formal complaint has been made.

The verified sources provided for this article do not include a direct response from OpenClaw’s maintainers to the reported similarities, or an independent technical comparison confirming the content matches. Nor do they document a formal complaint or related legal proceeding. This does not prove that no response or action exists outside these sources; it means that such a claim cannot be made on the material reviewed here.

The lack of detailed public documentation limits what can be concluded, but it does not make the reporting question irrelevant. A verifiable comparison could clarify whether the similarities are limited to naming conventions, include specific passages of text, or have another explanation. Until then, the acknowledgment of inspiration and the allegations about files should be kept in separate categories.

Readers following the story should watch for three developments: publication of a reproducible comparison of files from both products; a more precise explanation from Meta about the origin of the elements identified; and a direct response from OpenClaw’s maintainers. It would also help if any later usage figures clearly stated whether they measure installs, unique people, activity, or retention.

For now, the most accurate summary is that Muse attracted early attention and Meta acknowledged significant inspiration from OpenClaw, while news outlets reported similarities in certain files. The evidence in the available sources does not turn those facts and reports into a proven conclusion of copying or infringement.

For broader coverage of this launch and future updates, visit Inferama’s news section. To compare products and their capabilities, use the comparisons section; to explore other AI tools, browse the discovery section.

These are internal editorial links, not verification sources for the claims above.

Keep exploring

Corrections and transparency

If you spot incorrect or outdated information, send us a correction with the page and source we should review.