Omigo AI Companion App Review: Privacy Due Diligence

This review is a paper exercise, not a field test. Everything below comes from two documents captured on 22 October 2026: the Google Play listing for "Omigo: AI Character Chat" and the privacy policy published at the developer's own policy address. No account was created, no conversation was held, and no deletion request was filed. Where those two documents disagree or leave a gap, we say so rather than filling it. For a buyer weighing an emotionally intimate app, that gap analysis is more useful than a feature tour.

The buyer thesis, stated plainly

Someone considering Omigo is being asked to trust an app with some of the most sensitive text they will ever type: loneliness, fantasy, relationship talk, and possibly images of themselves. The reasonable due-diligence question is not "is the AI good?" but "do the public documents about this app hang together?" Three seams matter. First, the storefront labels the app Teen while policy Section 7 expressly prohibits minors and requires at least 18 plus local legal majority. Second, the name on the store listing is not the name in the policy. Third, the listing promises deletion can be requested, but a promise is not a mechanism. This review works each seam in turn and ends with a concrete decision rule.

What the store listing actually says, and what it does not

The Play listing identifies the app as "Omigo: AI Character Chat," shows "TOWN META" as the developer name, displays a content rating of Teen, and lists 50K+ downloads and in-app purchases. The description presents a broad character cast, mood-aware replies, a memory system, and image generation from text prompts. It also carries keywords such as roleplay and soulchat. All of this is developer-supplied marketing copy; it is a statement of intent, not evidence of how the app behaves on a given device. The listing was updated 19 October 2026, three days before capture, so treat the description as current at that date only.

Two structural details deserve attention. The content rating is a storefront classification, not a guarantee about the maturity of conversations a user can steer toward. And the "Teen" badge sits directly beside a policy that, as discussed below, contemplates adult companionship and payment features. That adjacency is the first seam.

The 18-plus question: reconciling the badge with the policy

The privacy policy does not describe Omigo as a children's product. It discusses account registration with date of birth, AI social companionship, membership services, and paid top-ups. It also references GDPR and CCPA, both of which impose heightened obligations around minors and require care about age assurance. The policy is explicit, not merely adult-sounding: Section 7 excludes minors and requires both an age of at least 18 and local legal majority. The storefront Teen classification does not override access conditions. This documentation audit did not examine how that restriction is enforced.

For a buyer, the practical reading is this: the store badge tells you what Google's rating system concluded from the developer's questionnaire, not what the operator's own legal posture is. If you are an adult seeking an AI companion, the mismatch should prompt you to check the in-app age confirmation yourself before paying for anything. If you are a parent, the Teen badge should not be read as reassurance. This is editorial advice, not a legal conclusion, and we make no claim about compliance with any age-verification regime.

Store identity versus policy operator: two names, one product?

The listing names TOWN META as developer. The policy, published 14 April 2026, states that the platform is operated by Blued Japan K.K. and uses "we" throughout. The policy URL sits on a micoportal.com subdomain, which is a third identity string again. A buyer is entitled to ask why three names appear around one product.

A plausible, ordinary explanation is a publishing entity (TOWN META) distinct from a policy-holding operator (Blued Japan K.K.), a pattern common when apps are distributed internationally. But the documents we captured do not say that. Nothing in the listing text we hold explicitly ties TOWN META to Blued Japan K.K., and nothing in the policy names TOWN META. That is a documentation gap, not proof of anything sinister. It does, however, mean that when you exercise a privacy right, you need to know which legal entity you are addressing and in which jurisdiction.

This distinction also affects deletion. A request sent to one entity is not automatically a request to the other unless they are the same controller. Before you rely on the policy's promises, identify the controller named in the app's own settings or support screens, and keep a record of where you sent your request.

A decision table for the three seams

Auditing the deletion promise without pretending to have tested it

The Play data-safety panel states that the developer declares no data shared with third parties, that data is encrypted in transit, that personal info, photos and videos, and device or other IDs may be collected, and that users can request deletion. Those are declarations. The policy adds texture: conversation history can be restarted or cleared through character settings, and memory information can be viewed and managed on a character profile.

Notice what is missing. Clearing a chat is not the same as deleting an account. Section 6.3 gives Profile > Settings > Delete Account and says deletion will be processed within 15 business days, or the legally required period, after identity and account-status verification. Associated information is to be deleted or anonymized unless law requires otherwise. The record does not demonstrate execution, a confirmation email or removal of training artifacts. It says de-identified data may be used for training and that re-identification is rendered impossible; it does not say deletion reaches derived memories or training artifacts.

Because we did not file a request, we cannot tell you whether deletion works. What we can tell you is how to audit it yourself cheaply. Create an account with a throwaway email, have one short conversation, request deletion through whatever channel the app names, and log the date, the channel, and any reply. If no channel is visible, that absence is itself the finding. Do this before you pour months of personal history into the product.

What the policy says about collection, in buyer terms

The policy describes collection of account data (username, date of birth where applicable, email or phone, profile fields), user inputs including text, voice, and images, interaction history, device and log data such as advertising identifiers and IP address, application lists and crash data, and payment records. It states that sensitive permissions like camera, microphone, and photo library are not enabled by default and are requested in context. It states that clipboard access may be used for share codes.

These are common patterns for a social app, and the policy is unusually explicit about them. The buyer-relevant point is scale: an AI companion app accumulates a longitudinal record of your emotional life, and the policy confirms that record is retained to provide continuity and may inform memory features. That is the product working as designed. Your job is to decide whether you want that record to exist, and under whose control.

Consumables, gems, and the economics you should not assume

Several dated reviews on the listing describe a gem or currency system that gates messages, with users reporting that free currency runs out quickly. One review from May 2026 describes a per-message cost and a daily earning ceiling; others from July 2026 describe running out mid-conversation. These are individual testimony at a point in time, not a price list, and we have not verified current mechanics. We deliberately do not repeat specific numbers as facts, because pricing and limits change and the reviews may not reflect today's build.

The structural lesson is that Omigo appears to be a consumable-driven product rather than an unlimited one. If your use case is long daily conversations with several characters, budget for ongoing spend and check the current in-app terms. The listing does show in-app purchases, which is consistent with that model but does not quantify it.

Hypothetical buying scenarios

These are illustrative only; no prices are asserted.

Hypothetical A: An adult user wants a private roleplay companion and plans to share personal history. They should first confirm the age gate, identify the controller in-app, and run the deletion audit above with a throwaway account. If any step fails, the privacy case is weaker than the marketing suggests.

Hypothetical B: A user mainly wants image generation from prompts. They should note the policy's statement that images and text are transmitted to servers for processing, and decide whether that is acceptable before uploading personal photos for avatars.

Hypothetical C: A parent sees the Teen badge and assumes suitability. Policy Section 7 expressly prohibits minors; the Teen badge does not override that restriction. Parents should not treat the store classification as permission to use the app.

Pros and cons from the documents

Pros. The policy is specific about data categories and permission timing, which is more than many companion apps disclose. The listing declares encryption in transit and no third-party data sharing. Memory and history controls are described as user-accessible. The app is actively updated.

Cons. The Teen content badge differs from the policy’s explicit 18-plus and legal-majority eligibility rule. Developer and operator names do not reconcile in the captured documents. Deletion has a documented in-app route and processing window, but neither was tested; retention exceptions and anonymized data require separate attention. Consumable economics are reported as restrictive by dated reviewers. None of these is disqualifying alone; together they justify caution before deep investment.

Precise questions to ask before you commit

1. Which legal entity is the data controller for my account, and in which country is it established? 2. What is the minimum age, and how is it enforced? 3. Is there a named deletion channel, and what is the stated response time? 4. Does deletion reach derived memories and any de-identified training data? 5. What does a message or image cost today, and what is the free allowance? 6. What happens to my data if I stop paying? Ask these in writing and keep the answers.

FAQ

Is Omigo only for adults? The policy is explicit, not merely adult-sounding: Section 7 excludes minors and requires both an age of at least 18 and local legal majority. The storefront Teen classification does not override access conditions. This documentation audit did not examine how that restriction is enforced.

Who owns Omigo? The listing names TOWN META; the policy names Blued Japan K.K. as operator. The captured documents do not link the two, so identify the controller in-app before sending rights requests.

Does Omigo sell my data? The Play data-safety panel declares no data shared with third parties. That is a developer declaration; read it alongside the policy's description of service providers and payment processors, and verify current terms.

Can I delete my data? The listing says you can request deletion. Policy Section 6.3 specifies Profile > Settings > Delete Account and processing within 15 business days after verification, or within the legally required period. It promises deletion or anonymization subject to legal exceptions; Section 4.2 permits indefinite retention of anonymized data. None of these outcomes was tested.

Is Omigo free? The listing shows in-app purchases, and dated reviews describe a consumable currency that limits messaging. We do not assert current prices or limits.

Is the memory feature reliable? The developer describes a memory system that retains preferences and past conversations. We have not tested recall accuracy and make no performance claim; treat memory as a described feature, not a verified one.

The decision

On the documents, Omigo is a plausible adult AI companion app with an unusually explicit privacy policy and an unresolved identity and age seam. The responsible decision is conditional: proceed only if you can answer the controller question and see a deletion channel before you share anything you would not want retained. Start with a throwaway account, run the deletion audit, and only then decide whether to invest conversationally or financially. If those steps fail, the privacy case does not support deep use, regardless of how good the chats are reported to be.