Effective from: 21/08/2026. Vistra uses artificial intelligence (AI) to help translate text on web pages and text inside images, right in the browser.
This English text is a translation provided for convenience. The Vietnamese version is the binding original; if the two differ in meaning, the Vietnamese version prevails. You can read it at Chính sách AI.
This policy explains how Vistra's AI features work, the limits you need to know about, and users' responsibilities when relying on AI-generated results.
1. Where AI is used
Vistra uses AI for features such as translating text on web pages, recognising and translating text inside images, and processing the content needed to render a translation onto an image.
These features run only when you actively request a translation.
Vistra may improve, extend or change how AI processing works in order to raise the speed, quality and stability of the Service.
2. How content is processed
When you request a translation, the necessary content may be sent to Vistra's systems and to the AI services needed to produce the result.
For text, the system processes the part that needs translating and returns the translation.
For images, the system may carry out steps such as recognising text, locating that text, understanding context, producing a translation, and preparing the data needed to display the translation over the image.
Vistra may change the technology or AI service used in order to maintain the quality, speed and availability of the Service. Data processing is carried out under the Privacy Policy.
3. Limits of AI
AI-generated results are not always accurate. AI may:
- mistranslate words or sentences;
- misread context;
- mistranslate proper nouns;
- get numbers wrong;
- misread handwriting;
- handle sound effects or special characters imprecisely;
- translate specialist terminology incorrectly;
- omit part of the content;
- place text incorrectly or off-position on an image;
- produce a translation worded differently from the original meaning.
Result quality also depends on the quality of the original content, image clarity, language, layout and context. You should check the original content where accuracy matters.
4. AI is not a substitute for professional advice
You should not rely entirely on Vistra's translations for content that could materially affect your rights, health, property or safety. For example: medical documents, medication instructions, legal documents, contracts, financial documents, transactions, safety instructions, or high-risk technical content.
Vistra's AI results are only intended to help users understand content and are not medical, legal, financial or other professional advice.
Where high accuracy is needed, you should check with an official source or a suitably qualified person.
5. Data and AI training
Content you ask to have translated is processed in order to provide the feature you requested.
Vistra does not proactively use your content to train AI models.
External AI services may process or store data under their own policies, security requirements and applicable rules.
Vistra may store some processing results, such as recognised text, translations, text positions or bounding regions, and the technical information needed to recognise content that has already been processed. These results may be reused when the system recognises the same content, in order to return results faster, reduce waiting time, avoid unnecessary repeated AI processing, and reduce operating costs.
Details are described in the Privacy Policy.
6. Third-party content
Vistra is a tool that helps translate content the user is accessing.
The fact that Vistra can translate a website, image, comic, document or other content does not mean Vistra owns that content or grants you rights to use it beyond what the law allows.
You are responsible for ensuring that accessing, translating, storing, using or sharing content is consistent with the law, intellectual property rights, the rights of the content owner, and the terms of the corresponding website or service.
Vistra does not encourage users to use AI tools to infringe the rights of other individuals or organisations.
7. Unsuitable results
AI can sometimes produce results that are wrong, confusing, out of context, contain unwanted wording, or differ from the meaning of the original content.
The appearance of an AI result does not mean Vistra endorses, agrees with or guarantees that content.
Vistra may apply technical measures or processing rules to limit harmful, unsuitable or unlawful results where necessary.
8. User responsibilities
You are responsible for how you use AI-generated results.
You may not use Vistra's AI features to carry out unlawful activity, fraud or harm to others, to infringe intellectual property, to infringe privacy, to exploit or attack systems, or to deliberately circumvent Vistra's quotas or protection mechanisms.
Vistra providing a translation tool does not mean Vistra takes responsibility on the user's behalf for how content is used after translation.
9. Changes to AI systems
AI technology changes quickly, and Vistra may upgrade models, change processing, change technical services, change how content is recognised or translated, adjust usage limits, or add or remove some AI features.
These changes may make the quality, speed or presentation of results differ from before.
If a change materially affects how personal data is processed, Vistra will update the Privacy Policy or provide notice where necessary.
10. Reporting AI results
If you encounter a wrong translation, a text recognition error, text placed in the wrong position, unsuitable content, or an AI feature fault, you can report it through the Feedback section in the App.
The information you provide helps us identify faults and improve the quality of the Service.
11. Contact
If you have questions or comments about Vistra's AI features, please contact: visiontranslate2026@gmail.com.