Tech Meridian ← LIVE FEED
PROMY MERIDIAN RU

RELEASE · COMPANIES · #3684

OpenAI releases CSAM image classifier, ChatGPT deterrent warnings, and developer guidance

OpenAI announced a new image classifier that detects child sexual abuse material (CSAM), including AI-generated images, and is providing select companies free API access; in tests it flagged 48% fewer images while identifying the same number of abusive images and produced flagged images twice as likely to be abusive. The company also added ChatGPT warning screens, temporary 24-hour account suspensions for attempts to access sexualized-child content, published developer guidelines for CSAM prevention/reporting, and rolled out law-enforcement tools, training, and a field guide, developed with partners including NCMEC, Tech Coalition, and others.

KEY POINTS

  1. OpenAI announced a new image classifier that detects child sexual abuse material (CSAM), including AI-generated images, and is providing select companies free API access; in tests it flagged 48% fewer images while identifying the same number of abusive images and produced flagged images twice as likely to be abusive.
  2. The company also added ChatGPT warning screens, temporary 24-hour account suspensions for attempts to access sexualized-child content, published developer guidelines for CSAM prevention/reporting, and rolled out law-enforcement tools, training, and a field guide, developed with partners including NCMEC, Tech Coalition, and others.
  3. The release provides tools aimed at reducing reviewer workload and improving detection of both real and AI-generated CSAM while expanding developer and law-enforcement capabilities to prevent and respond to abuse.

WHY IT MATTERS

The release provides tools aimed at reducing reviewer workload and improving detection of both real and AI-generated CSAM while expanding developer and law-enforcement capabilities to prevent and respond to abuse.

SOURCES & TIMELINE

1