OpenAI seeks to one-up Anthropic with new customer privacy protections
techcrunch ·
OpenAI is previewing a new privacy-focused safety service called Private Safety Processing, designed to monitor for misuse without retaining customer data. This comes as a direct contrast to Anthropic's recent 30-day data retention policy for covered models, which has concerned enterprise customers handling sensitive data. The new system uses automated agents to analyze inputs and outputs across multiple conversations, allowing it to detect malicious behavior that spans sessions without human review. If the system is triggered, it sends a narrowly defined signal to OpenAI, which may then contact the customer; customers can choose whether to share data. Anthropic, meanwhile, says any human review of customer data goes through a controlled path with tamper-proof logs. The move highlights the intense rivalry between the two AI companies, especially as a recent report showed OpenAI's Q2 grew more slowly than Anthropic, with Anthropic's annualized revenue run rate reportedly at $65 billion.
techcrunch