Would you use an AI “Money Leak Detector”?
I’ve been exploring a problem that many people experience but rarely notice clearly.
Most of us lose money in small ways like:
• forgotten subscriptions
• free trials converting to paid plans
• bills quietly increasing
• duplicate services
• credit card interest we could have avoided
• high credit utilization hurting our credit score
Individually these leaks seem small, but together they can easily add up to hundreds or even thousands per year.
So I started exploring an idea for an AI financial watchdog that detects these issues automatically.
Instead of focusing only on subscriptions, the idea is 360° money leak detection, including things like:
• trial-to-paid alerts before charges happen
• detecting duplicate subscriptions
• AI-assisted subscription cancellation
• alerts when a card exceeds safe utilization
• balance rebalancing suggestions
• guidance on which card to use for purchases
The goal is simple:
detect leaks early and suggest the next best action to fix them.
Before building anything real, I created a small prototype to test the idea.
Landing page:
https://mysentinel.one/
Curious about two things:
Would you actually use something like this?
Which financial “leak” annoys you the most?
Interesting angle and you're onto something real. The "death by
a thousand cuts" problem is well-documented but poorly solved
because most tools tackle one leak at a time (Rocket Money for
subs, Credit Karma for utilization) instead of a unified detection
layer.
Three thoughts that might sharpen your positioning:
-The "trial-to-paid before charge" alert is your strongest hook.
That's the leak people can feel they lost money on and get
genuinely angry about. I'd lead with that, not bury it in a
feature list.
-"AI-assisted subscription cancellation" is where most competitors
(Rocket Money, Bobby, Trim) actually make their money and lose user
trust, they take a cut or fail to cancel. Worth deciding early
whether you own that workflow or just alert and let users handle
cancellation themselves.
If you can tie it to "spend $X on card Y this week to optimize
your score before statement close," that's a very concrete, very
valuable feature.
To your questions:
subscriptions is a recurring frustration in our household.
that increase silently. Netflix going from $9 to $15 over 3 years
without anyone noticing
One question back: B2C consumer or B2B (employees tracking
corporate cards)? The answer dramatically changes your GTM.
Good luck. I subscribed to the waitlist
Thats great insight. Thank you so much for your feedback.
I am planning to start with B2C for now. Once the product is fully matured, can be extended to corporates.
Your feedback is much appreciated.
Early thought: I’m noticing credit card interest and utilization might be a bigger problem than subscriptions. Curious if others feel the same?
This is a really solid idea especially because most people feel like they’re losing money but don’t have visibility into where it’s happening.
I think the UX will be a huge differentiator here. If this feels like a passive dashboard, people might ignore it but if it behaves more like a proactive assistant (almost like ChatGPT-style nudges/notifications), it could actually change behavior. I think it will be important to be on all devices as well for maximum effectiveness.
The monthly summary idea could be powerful too: showing “money saved” vs “money still leaking” would make the value very tangible over time. This also shows the power of proactiveness, and usage of the tool instead of it joining the list of dozens of Ai tools that we don't actually use as much as we imagined.
One challenge that immediately comes to mind for me is trust. You’re asking users to connect sensitive financial data. Your ideal client is probably someone pretty serious about their finances. So how you communicate security and control will probably matter just as much as the features themselves. They need to know that the creators of the tool care a lot about this. "Privacy-first. Secure by design. Your data is never sold." Is a good phrase but it is tiny in comparison to how much the features are highlighted on your webpage. I know this is their speciality, but maybe check out nordvpn's wording about security. I think it can provide a couple of ideas.
Really cool idea!
Thank you for response.
You are right about the trust aspect. That is why we are considering methods which are industry standard and already securely established.
This is a really compelling problem space—especially because “money leaks” are usually not single big decisions, but lots of small, forgettable ones that compound over time. Framing it as a 360° detection layer (subscriptions, utilization, duplicate services, etc.) makes the idea feel broader and more useful than typical subscription trackers.
The biggest value here feels like timing + actionability—catching issues before they become charges or financial damage, rather than just reporting them after the fact. That shift is what could make this more than just another budgeting app.
I’ve seen similar financial behavior insights start getting traction when paired with AI-driven categorization and alerts in tools like ChatGPT-style assistants, where the real value comes from interpretation rather than raw data.
Curious—do you think users would trust an automated system enough to connect bank-level data, or would you position this more as a lightweight “recommendation layer” first?
Prize pool just opened at $0. Your odds are the best right now. If you're working on something new, this could help 👀
$19 entry → idea competition
🏆 Tokyo trip
💰 $500 guaranteed
Round open 👉 tokyolore.com
The 360° watchdog concept is solid—the trial-to-paid alerts are what usually gets me.
Regarding your validation: The best way to see if people actually use this is to see if they'll pay for it before the 'leak' even happens.
Since you've got the landing page live, you should throw MySentinel into the Validation Arena (tokyolore.com).
It’s a $19 sprint specifically for founders.
The winner gets a trip to Tokyo.
The prize pool is brand new ($0)