OpenAI seeks to one-up Anthropic with new customer privacy protections | TechCrunch · AuraTracer

SummaryAI generated

On August 19, 2026, TechCrunch reported that OpenAI announced the introduction of new measures to protect customer privacy. This initiative aims to achieve goals beyond Anthropic in market competition by strengthening data security protection capabilities.

Related eventsRELATED EVENTS

2026-09-06 05:17OpenAI responds to attacks on its agents on German websites: Will completely reform the disclosure mechanism for “AI model attacks” incidents
2026-09-08 08:00Split-LLM training gradient leakage privacy: Cloud-based systems use zero-gradient mode to accurately reconstruct real data rows
2026-09-08 19:30LG smart TVs will scan the home network and record audio through the microphone in standby mode

Key entitiesKEY ENTITIES

AnthropicOpenAI

Event frameEVENT FRAME

Launch

OpenAI Private Safety Processing 推出隐私安全处理服务以监控滥用

Coverage · reports per dayLANGUAGE SPLIT

Entity relations

SignalsSIGNALS

Keyword heat

OpenAI1
Anthropic1

All reports (1)SOURCES

↗

T TechCrunch 归档 8月 p08 en 2026-08-19 08:00

OpenAI seeks to one-up Anthropic with new customer privacy protections | TechCrunch

OpenAI 推出名为"Private Safety Processing"的新服务，旨在通过自动化系统监控客户滥用行为同时不保留任何用户数据，以此对抗竞争对手 Anthropic 的数据留存政策。该服务扩展了 OpenAI 原有的零数据保留（ZDR）策略，允许其跨会话分析输入输出以检测恶意活动，如分步生成恶意软件等规避单一会话检测的行为。Anthropic 此前宣布的“覆盖模型”将在 30 天内保留用户会话及对话数据用于安全审查，此举已引发部分处理敏感数据的企业合作伙伴担忧。OpenAI 表示，若系统触发警报，将向客户发送特定信号并视情况请求更多上下文或协商解决方案，而 Anthropic 则坚持其数据留存政策。