JAB AI

Does JAB AI have a moat?

No.

9/100 · Moatless

“A dozen AI boxing coach apps launched this year with the same voice rounds and video analysis. JAB AI's 483 ratings are its only head start.”

Code to copy

2 weeks

Moat to cross

3 months

Market

Crowded market

Share on X

Moat depth

MoatlessShallowReal

Revenue not verified. Founders can list on TrustMRR to show it here.

01/The verdict

A subscription mobile app ($4.99/week up to $79.99/year) for people training at home in boxing, MMA, Muay Thai, BJJ and 5 other martial arts. It gives AI workout plans, voice-called rounds, camera punch tracking and AI feedback on uploaded sparring videos.

Each part of JAB AI is easy to rebuild with off-the-shelf tools: LLM-written plans, text-to-speech combo callouts, phone pose tracking with something like MediaPipe, and video analysis sent to a vision model. The App Store already has many near-identical apps, including Boxer AI, Boxing AI, JabMaxx, ShadowCoach, two separate apps called Punch AI, BRAWLER and Braule, and someone even built one on Lovable. Its only lead is early App Store traction (483 ratings at 4.7 in the US, 10k+ Android downloads) and a small Discord. Paid ads and app store keyword work can catch up with that in months.

02/2 moats

Distribution

Shallow

~3 mo to cross

The US App Store listing shows 483 ratings at 4.7 stars, Google Play shows 10k+ downloads, and the Android listing title targets the 'Boxing & MMA Training' keywords. That likely helps it rank in app store search against newer clones.

To cross it: Spend on Apple Search Ads and TikTok creatives for a few months and push for in-app reviews until you have a few hundred ratings.

Switching costs

Shallow

~1 mo to cross

Version 2.1 keeps a history of rounds, meals, weight, a performance radar and weekly Arena leaderboards. Leaving means losing that history, but it is a few weeks of casual data at most.

To cross it: Offer a fresh start with a better first week of training. Most users of home fitness apps cancel within months anyway.

03/The gap

Easy to copy

Voice-called combos and round timers have existed in boxing timer apps for years
Video analysis that is likely a vision LLM call. The developer admits each video is judged on its own with no context from earlier uploads
Plans and tutorials for 9 disciplines that can be generated in bulk with AI
Fuel calorie tracking from meal photos, coach chat and streaks are standard features

How you'd go after it

Go deep on one discipline (BJJ or wrestling) with a real coach or gym brand behind the content, where a generalist AI's feedback falls short
Partner with a known fighter or a big combat sports YouTuber for distribution and credibility the solo developer lacks
Sell a white-label version to boxing and MMA gyms so members train at home under their own coach's brand
Undercut the $79.99/year plan with a generous free tier and win on accurate on-device punch and kick detection

04/Build against it?

Stay away

There are already a dozen AI boxing coach apps and the category gets new ones every month. Without a fighter or gym brand to bring customers, a new app is just another one in app store search.

Already selling this:Boxer AIJabMaxxShadowCoach BoxingPunch AIBrauleFightCamp

05/Receipts

Checked 2h ago by Claude Opus 5.5 on high effort, with web search. Confidence: medium.

Does yours?

Paste your website. You get your score in about a minute.