EN

Google Veo 3 - It's Worse Than You Think - Testing Demo & Review

#AI Tools
2025-06-06 20:498 min read

Content Introduction

In this video, the creator shares their frustrations with the V3 software and expresses disappointment over performance and customer service issues with Google. Following a failed refund request, they use their remaining credits to run V3 through a series of benchmark tests aimed at assessing prompt adherence and audio generation capabilities. The focus revolves around the lack of clear instructions on usage and the software's high costs compared to its performance. They provide statistical results of their tests, showcasing a low success rate and pointing out the complexities of generated content. Additionally, they intend to test alternative prompts created with the guidance of Gemini. They reflect on personal experiences, emphasizing the importance of clear instructions, and conclude by soliciting tips for using V3 effectively.

Key Information

  • The narrator discusses their experience with V3, emphasizing their disappointment with its performance and Google's refusal to issue a refund.
  • They conducted benchmark tests to measure V3's prompt adherence and audio generation capabilities.
  • Feedback from the audience prompted the narrator to address criticisms of their previous review and clarify their stance on V3.
  • The narrator seeks objective data to confirm V3's effectiveness through structured tests, promising to share results.
  • They explore generating prompts for V3 using another AI, Gemini, to test its capabilities, ultimately finding most prompts overly complex.
  • The narrator shares their results, noting a low success rate for usable content and highlighting their efforts to provide fair testing conditions.
  • They emphasize the need for better documentation and support from Google regarding V3's functionalities.

Timeline Analysis

Content Keywords

V3 Performance

The speaker discusses their experience with V3, using credits to run tests for its prompt adherence and audio generation capabilities. They express disappointment in its performance, sharing the lack of refunds and comparing interaction with Google to better alternatives like Midjourney.

Testing Methodology

The speaker outlines a series of tests conducted on V3 to evaluate its performance, mentioning the generation of prompts and audio. This includes detailed criteria for assessing generated content based on specific metrics.

Criticism of V3

The speaker addresses various criticisms received regarding their perspective on V3's functionality. They clarify the use of the term 'scam' based on their experience with Google regarding refunds, and the limitations of official guidance on V3.

Prompt Generation

V3's lack of effective prompt generation capabilities is highlighted. Tests reveal that prompts generated are often overly complex, leading to high production costs for usable content. The speaker emphasizes the importance of simplicity and effectiveness in generated outputs.

Video Results

The results of testing reveal that only a fraction of the generated videos are deemed usable. The speaker discusses the specific characteristics of the generated videos, such as repeated themes and visual effects, while also emphasizing the need for clear and straightforward prompts.

Audio Generation

The speaker critiques V3's audio capabilities, noting that a significant percentage of generated videos contain no audio or have distorted speech, leading to a low usability rate overall.

User Experience with AI

The speaker reflects on their history with different AI tools, sharing insights into how user guidance and robust documentation influence user satisfaction and success rates in AI-generated content.

Future of AI Video Creation

The speaker anticipates the future improvements in AI video generation and its potential affordability, discussing the associated costs in production and the importance of good prompt engineering for optimal results.

Community Engagement

The speaker encourages viewers to share insights and tips regarding the effective use of AI tools and prompt building to better utilize V3, emphasizing the need for community collaboration in navigating these technologies.

Related questions&answers

What is V3?

V3 is a software product that generates video and audio content based on user prompts.

Why did the author find V3 disappointing?

The author found V3 disappointing due to its performance and the lack of clear instructions or guidance from Google.

What kind of tests did the author decide to run on V3?

The author decided to run benchmark tests for prompt adherence and audio generation.

What was the success rate for the generated clips?

The author reported a 25% success rate, meaning two out of ten clips were fully usable.

How did the author determine the quality of V3's audio?

The author ran multiple prompts to generate audio clips and assessed their usability based on specific criteria.

What does the author suggest about Google's support for V3?

The author suggests that there is a lack of official support and documentation for V3 from Google, which impacts users’ ability to utilize the software effectively.

What was the author's experience with the Gemini AI?

The author interacted with Gemini to generate prompts for V3, but Gemini was found not to have knowledge of what V3 is.

How does the author feel about the pricing of V3?

The author expresses concerns about the cost relative to the value received, stating that the pricing seems high for the results observed.

What feedback did the author give on V3's performance?

The author found that V3 frequently generated complex and elaborate outputs that did not align with the project’s needs.

What does the author request from the audience?

The author requests links, best practices, or experiences from others on how to effectively use the scene builder feature in V3.

More video recommendations