Glacier AI Report soft launch
This is a special edition of Glacier's monthly blog, announcing the launch of our newest product, the AI Report. Glacier is also expanding its blog to add topics related to AI. For details and pricing on the AI Report, please reach out via email or use glaciernetwork.co/contact to request membership information.
OpenAI data preservation
On May 13th, 2025, a New York court ordered OpenAI to "preserve and segregate all output log data that would otherwise be deleted on a going forward basis until further order of the Court... whether such data might be deleted at a user’s request or because of ‘numerous privacy laws and regulations’ that might require OpenAI to do so." Very few people in data roles caught this order or saw the immediate impact it could have on their business. Fortunately, data from enterprise accounts (such as those using Azure) was not preserved in accordance with the order, primarily as a result of how those accounts happen to be managed. This court order demonstrates the speed at which business-critical issues can come up with popular AI tools, the difficulty of spotting and predicting them, and the lack of transparency around them.
Tech metaphors signal danger
It was informative to watch highly skilled attorneys litigate the details of OpenAI's data retention practices. The rules turned out to be more complex than perhaps most users would expect. The court even offered the metaphor of three gumball machines holding data from API, enterprise, and consumer accounts. So, gumballs featured prominently in the hearing. And in a bizarre coincidence, it so happens that (according to Circana) gum sales in 2024 were ~$3.5B, roughly the same as OpenAI's revenues. It may have been a more apt metaphor than we realized at the time.
Unknown unknowns
AI and the data sources that underly it are now developing so rapidly that complacency is perhaps the greatest threat to businesses. While Glacier was present at the subsequent NY court hearing on the OpenAI order, potentially impacted OpenAI users were mostly in the dark. Many were still looking into the issue weeks later, hoping for a response from a (likely overwhelmed) online support function or an account representative. That is an untenable position for businesses that integrate AI tools into both front and back offices, to say nothing of the fact that most vendors and service providers now also rely on these AI tools themselves, compounding the risks. A week is a long time to have one's proprietary data available to competitors and the public. Or to be unaware that this risk was on the table.
Take charge of AI risk
For more information about our AI Report product (gumballs not included), please email info@glaciernetwork.co or submit a membership request via the contact form at glaciernetwork.co/contact.
Don D'Amico
Founder & CEO, Glacier Network
©2025 Glacier Network LLC d/b/a Glacier Risk (“Glacier”). This post has been prepared by Glacier for informational purposes and is not legal, tax, or investment advice. This post is not intended to create, and receipt of it does not constitute, a lawyer-client relationship. This post was written by Don D’Amico without the use of generative AI tools. Don is the Founder & CEO of Glacier, a data risk company providing services to users of external and alternative data. Visit www.glaciernetwork.co to learn more.