Last updated
We build Architect, a competing AI journal. This review is built from Rosebud's own site, its App Store listing, press coverage, and public user reviews — not a hands-on scorecard. Every competitor fact is dated and linked. Last verified: October 12, 2026.
Looking for an alternative now that Rosebud has no free plan from September 30, 2026? We compared six.
You're comparing journals. Can we ask the harder question first — what would you write tonight if you were certain of who could hear it?
Hold that. First, the answer you actually came for.
↓ Skip to the answerRosebud Bloom is $12.99/month, or $107.99/year (billed annually — about $9/month). The iOS listing also carries heavier-use "Thrive" tiers from $24.99, sold only in the mobile apps. (rosebud.app + its App Store listing)
Not from September 30, 2026. Rosebud's help center says its free plan was retired effective that date; Bloom is $12.99 a month or $107.99 a year. Long-term memory is part of the paid plans. We couldn't find a published trial length on Rosebud's public pages, so we won't quote one.
Being met gently: morning and evening check-ins, therapist-designed CBT / IFS / ACT programs, habit tracking, voice journaling, polished iOS and Android apps.
Depth over time is the paid product. Memory — the thing the good reviews are actually praising — is on the paid plans, and its own help page says it recalls topics better than specific dates; by its own positioning, the AI is built to meet you rather than confront you, though a Challenger persona now exists; and its policy describes provider-side encryption, so the keys aren't in your custody.
The rest of this page is why.
Rosebud is, by wide reviewer consensus, one of the best-built AI journals you can open today. Two founders who met in a men's group, backing from Bessemer, and it shows: check-ins reviewers describe as designed rather than generated, therapist-built CBT, IFS, and ACT programs, voice journaling, weekly reports that notice your week for you. It now also offers five preset personas (the Sage, the Challenger, the Connector, the Gardener and the Strategist) plus your own, and works in 17 languages. If the hardest part of journaling has always been the blank page staring back, Rosebud removes the stare. That is not a small thing. Some nights it is the whole thing.
Rosebud publishes its own outcome research — survey responses, improvement within a week. Worth reading. Also worth knowing what it is: self-report after seven days, not a clinical trial. That's not a scandal; it's just the right size for the claim — and Rosebud deserves credit for publishing the underlying numbers at all. Almost nobody in this category does.
Two threads recur in public user reviews, and we'll attribute them rather than claim them first-hand: free-tier caps that could land mid-conversation, right when an entry was getting somewhere; and the discovery that the memory people praise — the app that "knows you" — is the paid product, not the free one. From September 30, 2026, the free plan itself is gone (its help center announced the change). Neither is a flaw exactly. Both are the business model showing through.
The worded verdict: a genuinely good product at what it sets out to do — a warm, well-crafted companion that lowers the cost of showing up. Whether it's built for what you need is the fork further down this page.
Not from September 30, 2026. Rosebud's help center says its free plan was retired effective that date. The paid Bloom plan is $12.99/month or $107.99/year, and long-term memory is part of the paid plans. We couldn't find a published trial length on Rosebud's public pages, so we won't quote one.
Rosebud details last verified: October 12, 2026
| Plan | Monthly | Annual | What it unlocks |
|---|---|---|---|
| Free | $0 until Sep 30, 2026 | — | No free plan from September 30, 2026 (its help center announced the change) |
| Bloom | $12.99 | $107.99 (≈$9/mo) | Long-term memory, deeper AI replies, full program library, weekly reports |
| Thrive 2x | $24.99 | $199.99 | Heavier daily AI usage (per the iOS listing) |
| Thrive 5x | $59.99 | $499.99 | Heaviest usage tier (per the iOS listing) |
Source: rosebud.app and its App Store listing · last verified October 12, 2026 (the App Store lists the Thrive prices without billing periods) · Rosebud also offers 40% off Bloom to eligible students, seniors and people with disabilities who subscribe on the web, and to service members and veterans — credit where due.
The honest math: the Rosebud you read about in the reviews — the one that remembers you — is the $12.99 Rosebud. From September 30, 2026, there is no free Rosebud to compare it with: budget for Bloom.
And since we're a competitor, here is our own card face-up next to theirs:
No free plan from September 30, 2026 (its help center announced the change). Bloom $12.99/mo or $107.99/yr; Thrive tiers for heavier use in the mobile apps. We couldn't find a published trial length. Last verified October 12, 2026.
Writing is free forever, no card. Your first 11 mentor responses are full power — a lifetime allowance, not a trial window. After that the mentor still replies once per rolling week, free. Full access is $33/month or $299/year ($24.92/month — saves $97), with a flexible $9.99/week option.
Yes, ours costs more. It's a different job — the sections below are the argument, and you're free to conclude the cheaper tool is the right one for you.
First, the thing cheap comparisons imply and we won't: nobody at Rosebud is reading your journal. Its privacy policy (updated September 8, 2026) says its broader staff has no access to production data, and that access to journal content is restricted to a small number of authorized people, behind two-factor authentication and audit logs, to fix a technical issue, answer a support request you made, or for safety and security. There's no reason to doubt it. That is not the question.
The question is architectural. The same policy describes provider-side encryption in transit and at rest — encrypted, genuinely, the way nearly every app you use works — and we found no client-side or zero-knowledge encryption claim in Rosebud's public materials. The policy is candid about why: keys held by the user carry a risk of losing your data if the key is forgotten or lost. That doesn't make it careless — it makes it typical of the category. It means your journal's confidentiality rests on policy and access controls — good ones, by all appearances — rather than on key custody.
Architect is built the other way around: entries are encrypted on your device with AES-256-GCM before sync, and the key never reaches us in readable form — it stays in your custody. The server stores ciphertext the company is mathematically unable to open. Not "we promise not to look" — cannot, by architecture.
And the honest fine print, because this page has been honest so far: when you ask the mentor to reply, that entry is processed transiently in plaintext to generate the reply — never stored readable, never used for training. Zero-knowledge at rest, transient processing for replies. Any vendor whose AI reads your journal has the same boundary; we'd just rather state ours than let you assume.
Why does this belong in a review of a gentle journaling app? Because of what the room does to the writing. What you're willing to write is the ceiling on what any AI can do for you. Gratitude lists and habit check-ins don't test that ceiling — Rosebud's architecture is perfectly fine for them. The thoughts you can't say out loud to anyone are the ones that need the room built differently. Feed any mentor the presentable draft and you get presentable advice.
Rosebud details last verified: October 12, 2026
Rosebud's reply is designed to meet you: warm, validating, therapist-informed. (It now also offers five preset personas, including one called the Challenger, if you want it to push.) That has real value — it's what sustains the habit, and in rough seasons it can be exactly the right thing. If that's what you came for, the verdict above stands: Rosebud does it about as well as anyone.
Architect's reply is designed to push: a Socratic mentor — four personas plus a custom one you design — that reads what you write and asks the question you were hoping nobody would. Clarity, not comfort. And the mechanic that makes the pushback land is memory: your history stays in play, and callbacks carry dates — when today's entry contradicts what you wrote six weeks ago, the mentor names the entry and holds the two up together. Your own words, quoted back at the right moment, land harder than anyone else's advice. It works in English, Turkish, and Spanish, in any good browser — excellent on phones, with no native store apps yet.
Neither reply is universally better. Warm reflection has roots in person-centered therapy; Socratic challenge has roots in cognitive therapy and most mentorship traditions worth the name. The honest routing: if you came to feel supported through a hard season, Rosebud's reply is the right one. If you came to change something specific about how you think, decide, or relate — and the same three problems keep coming back — the pushing kind is the one built for that job, and what an AI mentor actually does is closer to cross-examination than comfort. Neither app is therapy, and neither should claim to be.
| Architect | Rosebud | |
|---|---|---|
| Built for | Excavation — a private journal with an AI mentor that argues from your own history | Companionship — warm, therapist-informed check-ins that sustain the habit |
| Mentor posture | Socratic; pushes back — clarity, not comfort | Gentle and validating; tuned to meet you where you are (a Challenger persona is available) |
| Voices | 4 personas (The Architect, The Sage, The Mystic, The Billionaire) + a custom mentor you design | Five preset personas (the Sage, the Challenger, the Connector, the Gardener, the Strategist) plus your own, on mobile |
| Memory | Your history stays in play; dated callbacks cite the specific entry | Long-term memory on the paid plans; its help page says it recalls topics better than specific dates |
| Programs | No fixed curriculum — the mentor works from what you write | Therapist-designed CBT / IFS / ACT programs, habit tracking, weekly reports |
| Privacy | Zero-knowledge at rest (AES-256-GCM on device before sync; key never readable to us)* | Standard policy (provider-side encryption in transit and at rest); no client-side or zero-knowledge encryption claim in public materials |
| Voice features | Voice input, spoken playback, and the Sitting, a hands-free spoken session (beta) | Voice journaling with transcription, a hands-free mode, and replies read aloud |
| Languages | English, Turkish, Spanish — UI, mentor, and voice | 17 languages for the interface and the AI (its help center); voice journaling in 20 languages (its App Store listing) |
| Platforms | Web app (excellent in mobile browsers); no native store apps yet | Polished iOS and Android apps, plus web |
| Pricing | $33/mo or $299/yr ($24.92/mo — saves $97); flexible $9.99/wk | $12.99/mo or $107.99/yr (≈$9/mo); Thrive tiers from $24.99 in the mobile apps |
| Free tier | Writing free forever, no card; first 11 mentor responses at full power (lifetime), then 1 free reply per rolling week | No free plan from September 30, 2026 (its help center announced the change) |
* Encrypted where it's stored, and the key stays with you; a mentor reply processes that entry transiently in plaintext — never stored readable, never trained on. Details in the privacy section above.
Competitor facts last verified: October 12, 2026
You want to be met, not challenged — especially in grief, burnout, or a season where gentleness is the point. You want structure handed to you: therapist-designed programs, habit tracking, a check-in that's ready before you are. You want polished native apps today. And what you'd write isn't sensitive enough for key custody to matter. If that's you, budget for Bloom (from September 30, 2026 there is no free plan) — it's a well-made product.
You've been met, and the same three problems came back anyway. You want a mentor that reads what you write, remembers it with dates, and pushes — on the decision you keep re-litigating, the pattern you keep stepping around. You'd write more honestly knowing the stored entries are ciphertext even to the company. Or you think in Turkish or Spanish.
And they're not mutually exclusive jobs. Being held and being held accountable are both real needs; some seasons call for one, some for the other. The only wrong answer is pretending one tool does both.
The gentle reply keeps you writing. The pushing reply changes what you do next. Know which one you came for — then pick the room where you'd tell the truth.
Weighing the wider field of AI journals and mentors?