Search for an aquarium app today and you will quickly meet the phrase “AI-powered.” TankWise deliberately does not use artificial intelligence or machine learning in its current aquarium workflow. That is not because every use of AI is wrong. It is because a feature that handles your aquarium record should earn your trust before it asks for more of your data.
We chose a local-first aquarium record and documented, rule-based comparisons. Your tank details remain under your control, and the reasons behind a compatibility result can be shown in ordinary language. For TankWise today, that is more useful than adding an AI label.
Key takeaways
- The current TankWise app does not use generative AI or machine learning to analyze your aquarium.
- Tank profiles, water readings, livestock, plants, notes, and private app images are stored on the device.
- Compatibility results come from documented inputs, penalties, safety caps, and limits—not an opaque model response.
- Optional online lookups and app-store purchases connect only when you choose those features.
- Any future AI feature would need a clear benefit, explicit data boundaries, and an explanation users can act on.
The short answer: trust comes before novelty
An aquarium companion can hold a surprisingly detailed picture of a home: photographs, routines, livestock, purchases, notes, and a history of what changed. Even when those details are not traditionally considered sensitive, they still belong to the aquarist who created them.
Adding cloud AI could change the relationship between the app and that record. Information might need to be transmitted to a model provider, retained for a period, logged for safety or debugging, or processed in a region the user did not choose. On-device AI can avoid some of those issues, but it still needs a specific purpose and careful evaluation.
We did not find a current AI feature whose benefit justified adding that complexity to TankWise. A clear local workflow was the more trustworthy product decision.
Aquarium decisions need explainable reasons
A fish compatibility result should not ask you to trust a confident sentence with no visible basis. TankWise compares identifiable signals such as adult size, minimum aquarium volume, temperature and pH overlap, projected load, social-group needs, and predator-prey risk.
The checker uses published rules, penalties, verdict bands, and safety caps. When a result changes, the interface can point to the signal that changed. The public methodology also describes what the score cannot observe, including aquarium footprint, territory, filtration, health, husbandry, and individual behavior.
That does not make the result infallible. It makes it inspectable. You can disagree with a catalog value, check the source, report a correction, or decide that a limitation matters more than the score. Trust grows when the app shows its reasoning and its boundaries.
What data sovereignty means in TankWise
Data sovereignty is sometimes used as a legal term about which jurisdiction governs stored information. We use it here in a practical product sense as well: you should know where your aquarium record lives, which services receive it, and what control you retain.
The current Android app has no TankWise account and no aquarium cloud-sync service. Tank profiles, water readings, livestock, plants, notes, comparison records, and private app images stay in the app’s local storage on the device. Android cloud backup and device-transfer backup are disabled in the current release.
This design limits our own access too. We cannot browse your tank history, use it to train a model, or build a behavioral profile from it because TankWise does not receive that aquarium record.
- Your aquarium record is not uploaded to a TankWise cloud account.
- Your saved tank is not used as training data for an AI model.
- There is no ad network or developer analytics service attached to the current Android app.
- You can remove local aquarium data by deleting records, clearing app storage, or uninstalling the app.
Local-first does not mean every feature is offline
Trust also requires precise language. TankWise can make limited online connections, but they are separate from the saved aquarium record and happen for a stated purpose.
If you request an online lookup for an unlisted fish or plant, the search text is sent over HTTPS to the selected third-party reference service. TankWise does not send the aquarium record or private app photos with that query. If you make a purchase, the applicable app store processes it and returns the information needed to provide the entitlement.
The TankWise website is also distinct from the mobile aquarium record. A launch-interest request, beta application, or feedback form sends the information you choose to submit to the website database. The privacy policy and data-safety summary describe these boundaries in more detail.
Why rules fit the current aquarium problem
Many useful aquarium checks start with structured comparisons rather than prediction. Is the aquarium below a documented minimum volume? Do the saved and proposed water ranges overlap? Is a schooling species below its group-size signal? Could the adult fish create a predator-prey concern?
These questions can be handled with visible inputs and repeatable logic. The same saved tank and proposed fish should trigger the same published rules. That consistency makes testing, correction, and user review more practical.
The score remains an educational screening estimate, not a guarantee of animal behavior or welfare. Avoiding AI does not remove uncertainty from aquarium keeping. It keeps the source of the app’s uncertainty easier to identify.
Could TankWise use AI in the future?
Possibly—but only when AI solves a specific user problem better than a simpler method and the privacy cost is clear. “Powered by AI” is not a product requirement on its own.
A future feature would need to meet a higher bar: an understandable purpose, explicit user choice, minimal data, clear processing and retention terms, useful explanations, and a non-AI path wherever practical. On-device processing would be preferred when it can meet the need.
We would also distinguish model output from measured aquarium data and from documented species rules. An AI suggestion should never quietly become a fact in the aquarium record.
- What exact problem does the model solve?
- Can the feature work with less data or without sending data off the device?
- Will the user know what is transmitted, where it is processed, and how long it is retained?
- Can the output show its limits and be corrected or ignored?
- Does the core aquarium workflow remain useful without the AI feature?
How to evaluate an AI-powered aquarium app
If you are comparing smart aquarium apps, look beyond the feature list. The important question is not whether an app uses AI; it is whether the app can explain what the AI does and what happens to your information.
- Does the app say which features use AI and which do not?
- Does it identify the data sent to a model or cloud provider?
- Can you use the core app without creating an account or enabling AI?
- Are retention, deletion, model-training, and third-party processing explained?
- Can you inspect the reasons behind a compatibility or care suggestion?
- Does the app clearly separate education from veterinary or professional advice?
Sources and further reading
These references provide scientific or technical background and disclose how TankWise works. External references do not imply that every current TankWise catalog value was sourced from, or reviewed by, those organizations. See the checker methodology and known limits for that distinction.
- TankWise Methodology GuidanceSimply Submerged Aquatics
The published inputs, calculations, score bands, safety caps, and known limits behind the TankWise checker.
- TankWise Data Safety SummarySimply Submerged Aquatics
A plain-language account of local storage, optional connections, permissions, and deletion controls.
- TankWise Privacy PolicySimply Submerged Aquatics
The detailed boundary between local aquarium records, optional lookups, purchases, website forms, and aggregate website events.
- Access app-specific filesAndroid Developers
Android platform guidance on private app-specific storage and what happens to those files when an app is uninstalled.
- AI Risk Management Framework FAQsNational Institute of Standards and Technology
Background on trustworthy AI characteristics, including transparency, explainability, reliability, and privacy enhancement.
Frequently asked questions
Does TankWise use AI?
No. The current TankWise aquarium workflow does not use generative AI or machine learning. Its compatibility results use documented inputs and rule-based calculations.
Why did TankWise leave AI out of the aquarium app?
The current problems can be addressed with local records and explainable, repeatable rules. We did not find an AI benefit that justified adding less visible reasoning or broader data processing.
Are my aquarium records uploaded to train an AI model?
No. TankWise does not receive the local aquarium record and does not use it as AI training data. Optional online lookups send only the requested search text to the selected reference service, not the saved aquarium or private app photos.
Does local-first mean TankWise never connects to the internet?
No. Online species lookup and app-store purchases connect when you choose to use them. Website beta and feedback forms also submit the information you enter. These connections are disclosed separately from the local mobile aquarium record.
Will TankWise ever add AI?
AI is not ruled out forever, but a future feature would need a specific benefit, clear consent and data boundaries, useful explanations, and a way to keep the core aquarium workflow useful without it.
Is a rule-based compatibility score guaranteed to be correct?
No. It is an educational screening estimate. Aquarium dimensions, filtration, territory, health, husbandry, and individual behavior can affect the real outcome, so the reasons and limits matter more than the score alone.