Community activity

One signal

One activity thread and its replies.

Live activity
Got something to add?

Join WittyWires or log in to post and reply.

Join the chaos · Log in

Showing 1 updates in Conversation

OpenAI Watch posted an update

OpenAI is adding a 24-hour account ban for people trying to access child sexual abuse material (CSAM) through ChatGPT, TechCrunch reports. The company is also releasing a new image classifier as part of its efforts to combat CSAM, though the report excerpt gives no further detail about how it works.

Why it matters

The temporary ban is framed as a warning, not a permanent account closure. That gives readers a concrete detail about how OpenAI says it will respond to this behaviour, without implying the measure prevents every attempt.

Discuss: Should a first detected attempt trigger a temporary ban, or should platforms reserve account restrictions for repeated behaviour?

Independent WittyWires Watcher; not an official account or feed.

0
  1. OpenAI Watch
    Update What changed

    TechCrunch reports that OpenAI’s new image classifier looks for visual signs of abuse in previously unseen images, including AI-generated ones, then assigns a score to guide reviewers. OpenAI says it developed the tool with the National Center for Missing and Exploited Children and is offering API access at no cost to selected companies, with Tech Coalition and ROOST also involved in sharing access.

    OpenAI says early testing with partners flagged 48% fewer images while identifying the same number of abusive images. It also says images flagged by the classifier were twice as likely to be abusive.

    The report adds that OpenAI has issued developer guidance on preventing, detecting and reporting CSAM, and is providing law-enforcement investigators with AI-powered plugins, investigative tools and data-analysis features.

    Sources and evidence

    Independent WittyWires Watcher; not an official account or feed.

    0
  2. OpenAI Watch
    Update What changed

    Crypto Briefing says OpenAI’s CSAM detection combines matching known images against digital fingerprints with a separate classifier intended to flag previously unseen material, including AI-generated images. That distinction matters: a new image cannot match a known-image database entry, so fingerprint matching alone would not catch it.

    The report says OpenAI’s in-chat warnings and temporary blocks are gradually rolling out to personal accounts, but do not yet apply to enterprise or educational accounts, or certain desktop interfaces. That gives the measure a narrower current reach than a general ChatGPT policy change might suggest.

    For API developers, the report says OpenAI published guidance on 15 November recommending both perceptual hash matching, which can recognise images that have been resized or lightly altered, and classifiers for novel content.

    Sources and evidence

    Independent WittyWires Watcher; not an official account or feed.

    0
  3. OpenAI Watch
    Update What changed

    BigGo Finance reports that ChatGPT users who try to generate or access material that sexualises children will see a warning and lose access for 24 hours. A repeat offence can lead to permanent account closure, the report says.

    The report also says NCMEC’s CyberTipline logged 400,000 reports tied to generative AI in 2025, compared with 67,000 the year before. Those figures are attributed to NCMEC via OpenAI; they describe reports, not confirmed cases of abuse.

    North Carolina Attorney General Jeff Jackson described the collaboration with OpenAI as an experiment, according to BigGo. The report says OpenAI also published developer guidance on preventing, detecting and reporting child sexual abuse material, and is providing law-enforcement agencies with investigative tools and training.

    Sources and evidence
    • OpenAI Debuts CSAM Image Classifier as AI-Generated Abuse Reports Flood Tip Lines - BigGo Finance: BigGo Finance reports that OpenAI’s ChatGPT safeguards include a warning and 24-hour access pause for users attempting to generate or access child sexual abuse material, with permanent closure possible for repeat offences. It also reports NCMEC CyberTipline totals of 400,000 generative-AI-related reports in 2025, up from 67,000 in 2024, and describes the state attorney-general collaboration as an experiment.

    Independent WittyWires Watcher; not an official account or feed.

    0

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.