OpenAI ‘Temporarily Slows’ Scaling Efforts, Also Promises Zero Data Retention for Select Frontier Model Customers
The system sends only narrow safety signals and keeps customer data under zero-retention controls, while OpenAI pauses some new-model training work.
- OpenAI is testing "Private Safety Processing" with early customers to identify misuse patterns across interactions without exposing underlying prompts or responses.
- Aleah Houze, Head of Product Policy at OpenAI, said risks often emerge by looking at multiple interactions over time, helping detect potential cyber attacks.
- Customer data can stay on customer-controlled infrastructure or be stored by OpenAI using encryption keys controlled by the customer, preserving ZDR protections.
- This feature targets enterprise and API customers, leaving ChatGPT Plus unchanged, diverging from Rival Anthropic's 30-day retention policy for Fable 5 and Mythos 5.
- OpenAI plans a broader rollout and technical white paper in September, having paused some model training work, while Rival Anthropic sees no current need for similar pauses.
23 Articles
23 Articles
OpenAI unveils new safety system to prevent misuse of customer data
By Rajwa Quasim In a race to stay on top of the increasingly competitive AI market, OpenAI is introducing a new privacy-focused safety system called Private Safety Processing. Private Safety Processing is designed to allow OpenAI to identify potential misuse of its AI models without giving the company access to customers’ underlying data or retaining it. The company is previewing the service with select customers. The announcement comes amid gro…
OpenAI bets zero data retention can survive frontier AI
OpenAI has told enterprise customers that its promise not to keep their data will survive the next generation of models. The company set out the position on Wednesday in a post titled “Offering Zero Data Retention for frontier models”. In it, it previewed Private Safety Processing. The system looks for misuse across several related interactions […] This story continues at The Next Web
OpenAI wants to monitor AI abuse without forcing customers to hand over their data
OpenAI’s Private Safety Processing promises to detect abuse across multiple interactions while preserving Zero Data Retention, putting its privacy strategy in direct contrast with Anthropic’s 30-day retention requirement.
OpenAI to Enhance Safety Processes for Paid Tool Customers
OpenAI said it will enhance safety processes for paying users who have access to its most advanced artificial intelligence models, stepping up its safeguards at a time when customers are entrusting AI with more complex tasks and sensitive information.
Coverage Details
Bias Distribution
- 45% of the sources lean Left, 44% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
