Answer: On first-party services, Anthropic uses a hash-matching tool that calculates a perceptual hash of each image sent in an input and automatically compares it against NCMEC's database of known CSAM hash values; on a match, it notifies and provides NCMEC information about the input and the related Account, and also sends a notice to the user or organization any time it reports CSAM to NCMEC. Status: Confirmed Entry: tool/claude-apps/privacy-and-data/csam-detection-and-reporting Revision: 1 Digest: sha256:2ea48ef89726dfd7 (changes whenever the content changes) Timeline: Created: 2026-11-13T03:01:58Z Last changed: 2026-11-13T03:01:58Z Last verified: 2026-11-20T12:14:27Z Next check: 2026-11-21T12:14:27Z Check interval: Daily Expected change: No fixed schedule Data: Field | Value | Source | As of detection method | hash-matching tool | [1] | 2026-11-20 hash type | perceptual hash | [1] | 2026-11-20 database | NCMEC’s database of known CSAM hash values | [1] | 2026-11-20 reporting recipient | National Center for Missing and Exploited Children (NCMEC) | [1] | 2026-11-20 user notice | send a notice to the user or organization any time we report CSAM to NCMEC | [1] | 2026-11-20 page date | March 16, 2026 | [1] | 2026-11-20 Sources: [1] Anthropic. "CSAM Detection and Reporting | Claude Help Center". https://support.claude.com/en/articles/9020328-csam-detection-and-reporting Role: primary. Accessed: 2026-11-20T12:14:27Z Quote: "we use a hash-matching tool to detect and report known CSAM" (Supports: detection method) Quote: "we will calculate a perceptual hash of the image" (Supports: hash type) Quote: "This tool provides access to NCMEC’s database of known CSAM hash values." (Supports: database) Quote: "will report flagged media and related information to the National Center for Missing and Exploited Children (NCMEC)" (Supports: reporting recipient) Quote: "we will also send a notice to the user or organization any time we report CSAM to NCMEC" (Supports: user notice) Quote: "CSAM Detection and Reporting March 16, 2026" (Supports: page date) Snapshot SHA-256: 09d8218a7043326cc7e481d1c45063162ee2486a7961fd2bcff15858e15f69dc (captured 2026-11-13T03:01:58Z) Verification: Method: Automated extraction from [1] with deepseek-flash (JSON output, thinking disabled). Every value was checked by code to appear verbatim in a quoted passage of [1]. On this check the text of [1] was unchanged since 2026-11-13T03:01:58Z, so the values verified then were kept. Cross-check: not-available at 2026-11-20T12:14:27Z. No independent second source is configured for this entry Last successful check: 2026-11-20T12:14:27Z Last attempt: 2026-11-20T12:14:27Z Notes and limitations: Extracted automatically from a single official source (section "CSAM Detection and Reporting"). Related: https://latestn.com/tool/claude-apps/privacy-and-data/cvp-data-retention What are the data retention policies for the Claude Cyber Verification Program? https://latestn.com/tool/claude-apps/privacy-and-data/data-protection-when-improving-claude How does Claude protect my data when I allow my chats to be used to improve Claude? https://latestn.com/tool/claude-apps/privacy-and-data/data-sent-to-anthropic-from-inference Does Claude Desktop on a cloud provider send user data, prompts, or completions to Anthropic? https://latestn.com/tool/claude-apps/privacy-and-data/cowork-compliance-logging Is Claude Cowork activity captured in the Compliance API or OpenTelemetry? https://latestn.com/tool/claude-apps/privacy-and-data/cowork-where-tasks-run Where does Claude Cowork run my tasks and how does it access my files and apps? https://latestn.com/tool/claude-apps/privacy-and-data/crisis-conversation-sharing Does Anthropic send my Claude conversations to helplines or the safety plan? https://latestn.com/tool/ All entries in AI tools Change log: Time | Event | Change | Source 2026-11-13T03:01:58Z | added | Entry created | — Changes for this entry as a feed: https://latestn.com/tool/claude-apps/privacy-and-data/csam-detection-and-reporting/latest Challenge this entry: Open challenges: None How: If a value here is wrong or out of date, send what is wrong and a link to evidence. Challenges are queued and checked against the evidence; they do not change the entry by themselves. If the entry changes as a result, the change log says so. Form: https://latestn.com/dispute/tool/claude-apps/privacy-and-data/csam-detection-and-reporting POST: https://latestn.com/dispute/tool/claude-apps/privacy-and-data/csam-detection-and-reporting with claim, evidence_url, and optionally field and contact (form-encoded or JSON) GET: https://latestn.com/dispute/tool/claude-apps/privacy-and-data/csam-detection-and-reporting?claim={what is wrong}&evidence_url={link} How to cite: LatestN. (2026). How does Claude detect and report CSAM in user inputs? Revision 1, last verified 2026-11-20T12:14:27Z. https://latestn.com/tool/claude-apps/privacy-and-data/csam-detection-and-reporting (accessed 2026-11-21T10:07:19Z). @misc{LatestN_tool_claude_apps_privacy_and_data_csam_detection_and_reporting_r1, title = {How does Claude detect and report CSAM in user inputs?}, author = {{LatestN}}, year = {2026}, howpublished = {\url{https://latestn.com/tool/claude-apps/privacy-and-data/csam-detection-and-reporting}}, note = {Revision 1; digest sha256:2ea48ef89726dfd7; last verified 2026-11-20T12:14:27Z; accessed 2026-11-21T10:07:19Z} } Reuse: License: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/) Attribution: Copying and adapting are allowed, including commercially, if you credit LatestN and link to the entry you used. Each entry ends with a ready-made citation. URL: https://latestn.com/tool/claude-apps/privacy-and-data/csam-detection-and-reporting