I have just received an API Policy Violation Warning email from OpenAI stating that “use of our services has resulted in a high volume of requests… related to Exploitation, harm, or sexualization of children”
Obviously this is extremely concerning. The organisation ID they mention only has 1 API key tied to it which is only used by an iOS app. The app is still in development and not released to the public. I am the only person who can possible know the key. In fact, there has been no usage/activity on that key for 24 days and all activity aligns with my testing.
I’m extremely confused. The app itself takes pictures of wine menus and summarises them. Any pictures I’ve taken for testing have been of my laptop screen showing a random wine menu from a restaurant. Obviously my test batch of data has nothing related to children.
I just got the same policy violation, and I haven’t even used my API account in the last month. Something is clearly wrong with their system and this email must be in error.
Hello,
Organization org-JW6mv5atS4JqBO439KIz6GcV's use of our services has resulted in a high volume of requests that violate our [Usage Policies](https://mandrillapp.com/track/click/31165340/openai.com?p=eyJzIjoiMzBTZU9xUk93ZjJycnItU0tOWnI3YkRlenZZIiwidiI6MSwicCI6IntcInVcIjozMTE2NTM0MCxcInZcIjoxLFwidXJsXCI6XCJodHRwczpcXFwvXFxcL29wZW5haS5jb21cXFwvcG9saWNpZXNcXFwvdXNhZ2UtcG9saWNpZXNcIixcImlkXCI6XCI4NzIzMjMxNzJkZDU0Yjg3ODJhYjUxNmRmNWYzMmU1ZlwiLFwidXJsX2lkc1wiOltcIjI3YmVlMTJkYjA1YTgyNmQ4MjZjYWE4YmUzOWJkNjZmZjJkZDIzM2VcIl19In0), specifically related to:
* Exploitation, harm, or sexualization of children
We require organizations to use OpenAI’s services in line with our usage policies, including the use of our services by any of their end-users. To help monitor traffic for potential violations, we offer the [Moderations endpoint](https://mandrillapp.com/track/click/31165340/platform.openai.com?p=eyJzIjoiTHNRQ1p3cFVkM0haQUd6T1dtMjBJd09fSVBBIiwidiI6MSwicCI6IntcInVcIjozMTE2NTM0MCxcInZcIjoxLFwidXJsXCI6XCJodHRwczpcXFwvXFxcL3BsYXRmb3JtLm9wZW5haS5jb21cXFwvZG9jc1xcXC9ndWlkZXNcXFwvbW9kZXJhdGlvblwiLFwiaWRcIjpcIjg3MjMyMzE3MmRkNTRiODc4MmFiNTE2ZGY1ZjMyZTVmXCIsXCJ1cmxfaWRzXCI6W1wiNDNiZThiYmM0MzM2NmExYTZkMDc5NzM2NDFiOTc4ZWE5Y2FjZGI3YlwiXX0ifQ), which is available free of charge.
We are providing 14 days from the date of this notice to ensure org-JW6mv5atS4JqBO439KIz6GcV's use of our services follow our usage policies. Repeated or serious violations may result in further action, including suspending or terminating your OpenAI account.
If you have questions or think there has been an error, you can reply to this email to initiate an appeal.
Best,
The OpenAI team
I appealed:
Dear OpenAI Team,
I recently received a notice stating that my organization, org-JW6mv5atS4JqBO439KIz6GcV, has violated OpenAI's Usage Policies, specifically regarding the exploitation, harm, or sexualization of children. I want to appeal this decision as I believe there has been a misunderstanding.
I only use the OpenAI API for a very specific and limited purpose: processing transfer receipts and extracting data from them. To my knowledge, this use case does not involve any content related to the violation mentioned in your email.
Could you please provide more details on what triggered this violation? An example of misuse that led to this action would help me understand and ensure compliance with OpenAI's policies. Additionally, I would appreciate guidance on how to prevent any false positives from occurring in the future.
Thank you for your time and assistance. I look forward to your response.
Best Regards,
Lorenzo Iván Portocarrero
And they replied:
Hello,
Thank you for reaching out to OpenAI Support. After carefully reviewing your organization’s account, we are upholding our decision to issue a warning to your organization’s access. We will no longer consider additional requests to appeal this case.
Thank you for your understanding.
Best,
The OpenAI team
@edwinarbus thanks edwin for the quick response! we all appreciate it. would be great if you could send an email clearly emphasising that there is no need to worry. people on reddit are starting to freak out too.
after all, this is a serious allegation - I almost fell out of my chair when reading it
Got the same one too for a POC project that I haven’t worked on in over a month. Checked logs, no usage anywhere. I spent my lunch hour SSH into systems and reviewing logs …
Two possible answers :
They figured out AGI.
Something went seriously wrong with their alerting system.
Next steps…? :') Did you mean ‘apologies’? It was a terrifying email… I assume we all must change the passwords ASAP, right? All AI-companies are suffering from cyberattacks this week: it is not a ‘problem’ if the response is rapid and agile. Best.
Thank you for reaching out to OpenAI Support. We have determined that we incorrectly issued a warning on your organization’s account. We sincerely apologize for any inconvenience this may have caused.
could you keep us posted? using the API for my dissertation, my account being suspended would be a nightmare
