Lumo by Proton

Analysis by Norviv

I came away from Lumo by Proton with a fairly clear impression: it is most interesting when I treat it as a private thinking partner, not as a replacement for every productivity tool on my phone. That distinction matters. The app sits in the productivity category, but its value comes from helping me work with information through conversation rather than from offering a large collection of traditional office features.

Developed by Proton AG, Lumo is a free app with optional in-app purchases ranging from $12.99 to $119.88 per item. Its store summary describes it as a private AI assistant by Proton, and that privacy-focused identity shapes the experience more than a flashy collection of extras would. I can ask for explanations, reorganize rough ideas, draft text, or use it to think through a decision without opening several separate apps.

My overall view is positive, but measured. The app makes the most sense for people who want an AI assistant while keeping privacy near the top of their priorities. It is less convincing for anyone who mainly needs a polished document editor, a task manager, a research database, or highly specialized professional software. In those cases, a conventional app may still be the better tool.

What using Lumo feels like in everyday work

The first useful change is that Lumo gives me one place to start when my thoughts are still messy. Instead of deciding whether an idea belongs in notes, search, email, or a writing app, I can explain what I am trying to do in ordinary language and let the conversation help me shape it. That is especially handy at the beginning of a project, when the problem is not typing but knowing where to begin.

For example, imagine I have a busy afternoon and a half-written plan for a weekend trip. I can paste the rough notes into the conversation and ask for a clearer sequence, a shorter packing checklist, and a list of questions I still need to answer. I would still check the practical details myself, but the app can turn an unstructured block of text into something easier to act on. That kind of small transformation is where I found the assistant most useful.

I also like the possibility of asking follow-up questions without rebuilding the context from scratch each time. A normal search is often good at finding separate facts, while a note-taking app is good at storing what I already know. Lumo occupies the space between those activities: I can bring in my own material, ask for a different angle, and continue refining the result.

That does not mean every response deserves trust. I treat its answers as a first draft, an explanation to inspect, or a prompt for further thinking. When the subject involves money, health, legal obligations, current events, or an important professional decision, I verify the result elsewhere. The convenience is real, but convenience should not be confused with authority.

The privacy-first angle changes the buying decision

Privacy is not just a decorative label here. It is the main reason I would choose this assistant over a more familiar general-purpose AI product. Proton AG has built its reputation around privacy-oriented services, so people who already value that approach will probably understand the appeal immediately. For me, the attraction is being able to use an AI workflow without treating the most popular option as the automatic choice.

Still, privacy-minded users should keep their own habits sensible. I would not paste passwords, private access codes, confidential client material, or personal documents into any conversational assistant simply because the product presents itself as private. The safest workflow is to remove identifying details first, especially when the task can be completed with a simplified example.

A useful trick is to separate the content from the identity around it. If I want help rewriting a sensitive message, I can replace names, account numbers, and locations with neutral labels. Lumo can still help with tone and structure, while I reduce the amount of unnecessary personal information involved. That is a more practical privacy habit than assuming the app can remove every risk for me.

Where the assistant saves real time

I found three workflows more valuable than casual question-and-answer use. The first is compression: turning a long passage into a short briefing, a checklist, or a set of action points. The second is transformation: changing a rough message into something warmer, firmer, clearer, or more professional. The third is comparison: asking the assistant to organize competing options according to criteria I provide.

The comparison workflow deserves a little care. If I ask which of two choices is “better” without explaining what matters to me, the answer tends to be less useful than when I define the trade-off. I get better results by saying whether I care most about cost, simplicity, flexibility, time, or privacy. Lumo is more helpful when I give it a decision framework instead of asking it to invent one.

Another practical technique is to request a response in stages. I might first ask for the main issues, then ask it to challenge its own recommendation, and finally ask for a short action plan. This reduces the temptation to accept the first polished answer. It also makes the conversation easier to audit because I can see how the conclusion developed.

For writing, I would use it as an editor rather than as an invisible author. I can provide my own points and ask for a clearer order, remove repetition, or adjust the tone. That preserves more of my voice than requesting a complete article from a blank screen. It also gives me something concrete to judge: whether the revised version still says what I actually mean.

What it does not replace

Lumo is not a substitute for a dedicated calendar, a full task manager, or a serious word processor. A conversation can help me design a weekly plan, but it is not the same as having recurring reminders, due dates, sorting controls, and a dependable visual overview. Likewise, an assistant can draft an email, but sending, tracking, and organizing that email belong to another tool.

This difference is important because AI apps can appear to cover everything at first glance. In practice, they often help with the thinking around a task rather than the complete operational task itself. I would use Lumo to decide how to structure a project and then move the final actions into the app designed to manage them.

The same caution applies to research. Lumo can help me understand a topic, identify subquestions, and make dense material easier to read. It should not become my only source for claims that need current or authoritative verification. For serious research, I still want original documents, reputable references, and a way to inspect where important information came from.

Small frictions that affect the experience

The conversational format is both the app’s strength and its main limitation. When I know exactly which button I need, a specialist app is usually faster. A calculator, a checklist, or a document template can be more direct than explaining the desired result in a chat. Lumo earns its place when the task is ambiguous, text-heavy, or open to interpretation.

There is also a natural risk of conversation drift. After several follow-up questions, the discussion can become harder to manage, particularly when I change direction halfway through. I get better results by starting a fresh conversation for a new subject and by restating the goal when the task becomes complicated. That simple habit prevents old assumptions from quietly influencing a new request.

Another weakness is the polished-answer problem. A response can sound organized while still missing an important exception or making an unsupported assumption. I pay particular attention when the assistant summarizes material I did not write myself. A concise answer is useful only if the compression has not removed the qualification that changes the meaning.

The current version is 2.0.5-gms, and the app requires Android 10 or later. That makes it unsuitable for people using older Android devices, even if the phone still handles their other productivity apps comfortably. The age rating is Everyone, which fits the general-purpose nature of the tool, but an Everyone rating should not be read as advice to share every kind of personal information with it.

Who will get the most from it

I would recommend Lumo first to privacy-conscious students, writers, independent workers, and ordinary users who want help turning scattered thoughts into usable text. It is particularly good for people who already know how to check AI output and who are willing to guide the assistant with clear instructions.

Students can use it to break a difficult subject into smaller questions, create a revision outline, or compare explanations written at different levels of complexity. I would not use it to bypass learning. A better approach is to ask for an explanation, attempt the problem independently, and then use the assistant to identify gaps in the reasoning.

Writers may appreciate the editing workflow. Instead of asking for generic prose, I can supply a draft and request specific changes: shorten the introduction, remove repeated points, make the tone less formal, or identify where a reader may become confused. The more precise the request, the more control I retain over the result.

People who already use Proton services may also find the product easier to justify because the privacy-centered ecosystem is familiar. The free entry point makes it straightforward to try before deciding whether the optional purchases provide enough value for a heavier routine. I would begin with occasional tasks rather than committing to a paid plan immediately.

Who should choose another kind of app

I would steer away from Lumo if the main goal is strict project management. Someone who needs dependable task dependencies, team assignments, detailed reporting, or a carefully structured database will be better served by a dedicated productivity platform. An AI conversation can support that work, but it should not be mistaken for the underlying system.

I would also hesitate to recommend it to users who want automatic, hands-off answers. The best experience requires judgment: writing a useful prompt, supplying enough context, spotting weak assumptions, and checking consequential claims. If that sounds like extra work, a simpler specialist app may feel more satisfying.

People with older Android hardware should check compatibility before planning around it, since Android 10 is the minimum operating system. And anyone choosing primarily on average rating should know that the app currently sits at a 3.7 average from around 2.9 thousand ratings, with 285 written reviews. That is a mixed reception rather than a universal endorsement, so I would approach it as a tool worth testing for a particular workflow, not as an automatic upgrade to every phone.

Value, availability, and the sensible way to try it

The app is free to install, which lowers the barrier for experimenting with it. Optional purchases are listed from $12.99 to $119.88 per item, so the financial decision depends on how often I would use the assistant and whether its privacy-oriented approach matters enough to justify spending beyond the free experience.

My advice is to test it with three real tasks before paying: one writing task, one organization task, and one explanation or comparison task. Use material that is safe to share, note whether the answers actually reduce your workload, and see whether you prefer the conversational approach to your existing tools. This is more revealing than asking a few entertaining questions and judging the app from those alone.

The app was released on Jun 6, 2025, and has passed 100 thousand installs. Those figures suggest that it has found an audience, but they do not settle the personal question of whether it belongs in my daily routine. For an assistant, fit matters more than popularity because the value depends heavily on the kinds of tasks I bring to it.

My final recommendation

Lumo by Proton is a thoughtful productivity app with a clear identity. Its strongest quality is not that it can produce fluent text; many assistants can do that. Its stronger case is the combination of conversational help and Proton AG’s privacy-focused positioning, giving me a reason to consider it when I do not want the most obvious mainstream choice.

I would keep it on my phone for outlining, rewriting, summarizing, brainstorming, and working through decisions that benefit from a second perspective. I would not make it the only place where I manage deadlines, store sensitive records, verify important facts, or run a serious research process.

My recommendation is simple: try it if private AI assistance is more important to you than having a complete all-in-one productivity suite. It is a good fit for reflective, text-based work and a weaker fit for rigid workflows that demand dedicated controls. Used with that boundary in mind, it feels useful rather than overpromised, and that is exactly the kind of assistant I would rather recommend to a friend.