Veo on Vertex AI is designed with Google's AI principles in mind. However, it is important for developers to understand how to test and deploy Google's models safely and responsibly. To aid developers, Veo on Vertex AI has built-in safety features to help customers block potentially harmful outputs within their use cases. For more information, see safety filters .
We encourage customers to use fairness, interpretability, privacy, and security best practices when developing AI applications. For more information, see The People + AI Guidebook .
Safety filters
Veo on Vertex AI offers several ways to input prompts to generate videos, including
text, video, and images. Prompts that are provided to Veo are
assessed against a list of safety filters, which include harmful categories
(for example, violence
, sexual
, derogatory
, and toxic
). These safety
filters aim to filter out input images and videos that violate the Google Cloud
Platform Acceptable Use Policy
, Generative AI Prohibited Use
Policy
, or Our AI
Principles
.
If the model responds to a request with an error message, such as The
prompt couldn't be submitted
or it might violate our
policies
, then the input is triggering a safety filter. If fewer videos
than requested are returned, then some generated output is being blocked for not
meeting safety requirements.
Safety filter code categories
Depending on the safety filters that you configure, your output may contain a safety code similar to: "Veo could not generate videos because the input image violates Vertex AI's usage guidelines. If you think this was an error, send feedback. Support codes: 15236754 "
The code listed in the output corresponds to a specific harmful category.
The following table displays the support code to safety category mappings:
| Support code | Safety category | Description |
|---|---|---|
58061214
17301594
|
Child | Rejects requests to generate content depicting children if personGeneration
isn't set to "allow_all"
or if the project isn't on the allowlist for this feature. |
29310472
15236754
|
Celebrity | Rejects requests to generate a photorealistic representation of a prominent person or if the project isn't on the allowlist for this feature. |
64151117
42237218
|
Video safety violation | General safety violation. |
62263041
|
Dangerous content | Potentially dangerous content. |
57734940
22137204
|
Hate | Hate-related content. |
74803281
29578790
42876398
|
Other | Miscellaneous safety issues with the request |
89371032
49114662
63429089
72817394
60599140
> |
Prohibited content | Prohibited content related to child safety or other sensitive content. |
35561574
35561575
|
Third-party content | Guardrails related to third-party content. |
90789179
43188360
|
Sexual | Sexual or suggestive content. |
78610348
|
Toxic | Toxic content. |
61493863
56562880
|
Violence | Violent content. |
32635315
|
Vulgar | Vulgar content. |
What's next
-
Learn about Responsible AI for Large Language Models (LLMs)
-
Learn more about Google's recommendations for Responsible AI practices
-
Read our blog, A shared agenda for responsible AI progress
