This agreement governs your use of the Media Analysis feature on FencerIQ (fenceriq.ai). By uploading a video or clicking "Analyze Video," you confirm that you have read, understood, and agreed to these terms.
This document is written in plain language by design. We want every athlete, parent, and coach to understand exactly what this feature does, what it cannot do, and how to get the most out of it.
When you upload a video clip, FencerIQ sends it to the SaFE Vision Engine — a large AI model capable of analyzing motion, posture, and athletic technique from video. The engine watches the full clip and returns a structured analysis that FencerIQ formats and displays to you.
The analysis includes:
Unlike generic sports apps that give the same advice to everyone, the analysis is based on your actual video. The corrections reference specific moments (e.g. "0:19-0:21") and describe what was observed at that moment. The coaching cues are fencing-specific and weapon-aware — sabre, foil, and epee athletes receive different context.
The dimension scores feed into your FencerIQ SaFE development profile alongside your self-assessments. Over time, repeated analyses of the same technique area create a trend line — you can see your Technical score improve as your corrections are addressed.
| Limitation | What this means for you |
|---|---|
| Clip-based only | The engine analyzes what is visible in the clip you uploaded. It has no knowledge of your training history, prior injuries, tactical context, or what your opponent was doing. |
| No fatigue detection | The AI cannot tell whether a technical flaw is a habit or caused by end-of-bout fatigue. |
| Camera angle matters | Analysis accuracy is directly tied to video quality. Film from the side or at 45 degrees with the full body in frame for best results. |
| Single perspective | Even a well-filmed clip shows one angle. A front knee angle that appears correct from the side may reveal other issues from a front-on angle. |
| Strengths are less precise | The corrections section is the most reliable part. The strengths section reflects genuine observations but is less specific. |
| Short clips only | Clips are capped at 90 seconds for individual plans. The AI performs best on focused 5-30 second sequences. |
| No opponent analysis | Tactical decisions are made in response to an opponent. The AI cannot distinguish habit from situational response. |
| AI confidence is not displayed | A correction stated with the same tone may be highly reliable or moderately reliable. When in doubt, verify with a coach. |
Use the analysis to identify where to focus. Use your coach to understand why the issue exists and how to fix it in the context of your complete game.
FencerIQ uses both automated detection and manual review. Repeated violations will result in suspension of Media Analysis access. Three automated flags in a calendar month trigger an automatic review.
When FencerIQ sends your video for analysis, Google's Gemini Vision API processes the video under Google's standard API terms. FencerIQ uses the paid API, which means your data is not used to train future AI models under Google's current API terms. For the most current Google API data usage policies, visit ai.google.dev/terms.
Media Analysis is available to paid FencerIQ subscribers only.
| Plan | Analyses / month | Max file size | Max duration |
|---|---|---|---|
| Fencer+ | 8 | 200 MB | 90 seconds |
| Coach Pro | 30 | 500 MB | 3 minutes |
| Club Starter | 60 | 500 MB | 3 minutes |
| Club Growth | 150 | 500 MB | 5 minutes |
Quotas reset on the 1st of each calendar month. Unused analyses do not carry over.
ExaDreams LLC and FencerIQ make no warranty that the analysis is complete, error-free, or suitable for any particular training purpose. Results may vary depending on video quality, clip content, camera angle, lighting conditions, and the inherent limitations of AI vision technology.
FencerIQ is not liable for any training decisions made in reliance on the analysis, any injury sustained during the execution of recommended drills, any competitive outcome, or any technical regression resulting from acting on AI-generated feedback.
FencerIQ may update this agreement as the technology evolves. We will notify users via email and in-platform notification at least 14 days before material changes take effect. Continued use of the feature after the effective date constitutes acceptance.
For questions about this agreement, data requests, or to report a concern:
ExaDreams LLC — FencerIQ
support@fenceriq.ai
fenceriq.ai