Editing Openai/692c4a6d-a0b8-8002-af18-05564d5876c3


{{DISPLAYTITLE:Yuh AI policy red flags}}

''This conversation was exported from ChatGPT''<ref>{{cite web|title=Yuh AI policy red flags|url=https://chatgpt.com/share/692c4a6d-a0b8-8002-af18-05564d5876c3|publisher=ChatGPT|access-date=2025-11-30}}</ref>

== gpt-5-1 ==
=== User: Can you tell me if there are any red flags in these new policies regarding AI use in the Yuh App ===
Can you tell me if there are any red flags in these new policies regarding AI use in the Yuh App

=== Assistant: Below is a clear and direct assessment of red flags specifically related to AI use, based solely on the three documents… ===
Below is a clear and direct assessment of red flags specifically related to AI use, based solely on the three documents you provided:
* Yuh App Terms & Conditions terms-and-condititions-yuh-app-…
* Yuh Account Privacy Policy (Swissquote) yuh-account-privacy-policy-d-en…
* Yuh App & Website Privacy Policy (Yuh Ltd.) yuh-app-privacy-policy-en-v3

=== Major AI-Related Red Flags ===

==== 1. Your chatbot conversations may be reviewed by humans ====

The Yuh App Terms explicitly state that chatbot conversations may be stored and used for analysis including manual human review to improve services.
terms-and-condititions-yuh-app-…

Why this matters:
This is not standard for banking apps where user conversations can contain sensitive financial intent, account-related details, etc.

==== 2. Your data—including chatbot inputs—can be transferred outside Switzerland, including to the US ====

The Terms state that personal data sent to the chatbot can be transferred to jurisdictions without an equivalent level of data protection, explicitly naming the United States.
terms-and-condititions-yuh-app-…

Both privacy policies confirm international transfers for many purposes, including analytics, payment processing, marketing, and more.
yuh-account-privacy-policy-d-en…

 yuh-app-privacy-policy-en-v3

Why this matters:
US-based providers may be subject to broad federal access rules (e.g., the CLOUD Act).
Your banking-related conversations could therefore lose Swiss-level confidentiality.

==== 3. You explicitly waive Swiss banking secrecy in several cases ====

The Swissquote privacy policy includes a broad banking secrecy waiver, including for:
* AI-powered tools
* foreign third-party processors
* card processors (e.g., Marqeta in the US) yuh-account-privacy-policy-d-en…

Why this matters:
This is unusually broad for a Swiss financial institution.
Your data is no longer protected under strict Swiss secrecy rules once it leaves Switzerland.

==== 4. AI output is disclaimed heavily (“hallucinations,” no reliability, no liability) ====

The Terms emphasize that the chatbot may produce:
* inaccurate
* incomplete
* misleading information, including hallucinations. terms-and-condititions-yuh-app-…

Why this matters:
If you rely on chatbot responses for financial decisions, they disavow all responsibility.

==== 5. Chatbot data is used to train or improve models (not just to answer you) ====

The Terms state the data may be used to:
* improve the chatbot
* develop new products
* improve the Yuh App terms-and-condititions-yuh-app-…

Why this matters:
Your personal queries, habits, and metadata can feed internal AI training pipelines.

==== 6. Profiling is explicitly permitted ====

Both privacy policies describe profiling, including assessing aspects of your personality, especially for marketing.
yuh-account-privacy-policy-d-en…

yuh-app-privacy-policy-en-v3

Why this matters:
Your interactions—possibly including AI chatbot conversations—may influence personalized marketing, risk scoring, or eligibility assessments.

=== Moderate Red Flags ===

==== 7. Data shared with many third parties, including Big Tech analytics providers ====

Yuh and Swissquote share data with:
* Google (analytics, cookies)
* Apple/Google Push services
* Marqeta (US), Mastercard, payment partners
* Other analytics tools yuh-app-privacy-policy-en-v3

Push notifications can expose sensitive data and pass through foreign servers.
terms-and-condititions-yuh-app-…

==== 8. No promise that AI data will stay encrypted end-to-end ====

While they say data is encrypted “to some extent,” they clearly acknowledge:
* data may travel uncontrolled across foreign networks
* even encrypted data may expose who is communicating yuh-account-privacy-policy-d-en… yuh-app-privacy-policy-en-v3

==== 9. Exercise of your rights may result in loss of service ====

If you object to certain processing (including AI-related profiling or data transfer), they may:
* stop providing the service
* terminate your access yuh-account-privacy-policy-d-en… yuh-app-privacy-policy-en-v3

Why this matters:
There is effectively no meaningful opt-out for AI-related tracking or profiling.

=== Low-Severity (but notable) AI-related concerns ===

==== 10. The AI chatbot can influence your financial behaviour without accountability ====

They emphasize the chatbot’s information is:
* not advice
* not recommendations
* not reliable terms-and-condititions-yuh-app-…

But since users may treat it as advice, this offloads all responsibility.

==== 11. Conversations can be linked to your account identity ====

Because the chatbot is inside the banking app, conversations are necessarily tied to:
* device identifiers
* account data
* customer profile terms-and-condititions-yuh-app-…

This is substantially different from an anonymous chatbot.

=== Summary — The Biggest Red Flags ===

Highest-risk items:
# Human review of your chatbot conversations
# Data transfers outside Switzerland (incl. US)
# Waiver of Swiss banking secrecy
# Use of your data for training/improving the AI
# Profiling for marketing and behavioral analysis
# No reliable AI output, no liability

If you're concerned about privacy, confidentiality, or regulatory compliance, these are non-trivial red flags.

If you want, I can create a short risk-rating table, suggest a safe way to continue using the app, or draft questions to ask Yuh's Data Protection Officer.

== References ==
<references />


[[Category:ChatGPT Shares]]
[[Category:openai]]