Subject: Feature Request: Adding Positive Feedback Mechanism to Google Veo/Flow Interface
Dear Google Veo Development Team,
I’m writing as an active user of Google Veo through the Flow interface who has been exploring its capabilities for commercial content creation. While I appreciate the current feedback system that allows us to flag issues with generated content, I’ve noticed a significant gap that I believe impacts both user experience and your ability to improve the model effectively.
Currently, the interface only provides a “flag” or “report issue” option for each generation. This creates an asymmetric feedback loop where you’re only capturing data about what goes wrong, missing crucial information about what works exceptionally well. As someone who has worked extensively with Veo, I’ve encountered numerous instances where the AI has exceeded my expectations or perfectly captured my creative vision, but I have no simple way to communicate this success back to you.
The absence of a positive feedback mechanism creates several challenges:
1. **Incomplete Training Data**: Without positive signals, your model improvements are based on an inherently biased dataset that only tells you what to avoid, not what to reinforce. This is like teaching a cinematographer by only pointing out mistakes without ever acknowledging when they nail the perfect shot.
2. **User Frustration**: When Veo produces exceptional results, users naturally want to acknowledge this success. The current system forces us to either remain silent or use cumbersome workarounds like taking screenshots for community forums or waiting for user research invitations.
3. **Missed Learning Opportunities**: Some of my most successful generations have come from experimental prompts or unexpected parameter combinations. Without a quick way to mark these successes, valuable insights about effective prompting techniques may be lost.
4. **Community Building**: A positive feedback option would help identify exemplary outputs that could be shared (with user permission) as learning resources for the broader community, accelerating everyone’s mastery of the tool.
I propose adding a simple “thumbs up” or “mark as excellent” icon alongside the existing flag option for each generation. This would provide several benefits:
– Give your team balanced data about model performance
– Allow users to quickly signal when Veo meets or exceeds expectations
– Help identify patterns in successful generations that could inform model improvements
– Create a more positive and engaging user experience
– Enable you to build a corpus of “gold standard” examples for different use cases
Additionally, when users click the positive feedback icon, you could optionally present a quick form asking what specifically worked well (e.g., “accurate camera movement,” “perfect lighting,” “maintained consistency,” etc.), similar to how the flag system might ask what went wrong.
This feature would be particularly valuable for professional users like myself who are evaluating Veo for commercial applications. Being able to systematically track both successes and failures would help us build more reliable workflows and provide you with higher-quality feedback about real-world use cases.
I understand that implementing new features requires careful consideration and development resources. However, I believe this relatively simple addition would significantly enhance the value of user feedback and accelerate Veo’s improvement trajectory.
Thank you for considering this suggestion. I’m happy to participate in any user research or beta testing related to feedback mechanisms, as I’m invested in helping make Veo the best tool possible for creative professionals.
Best regards,
Kenn Jordan
CEO of ElmsPark
info@elmspark.com