Answer: The text does not list specific ChatGPT features labeled Elevated Risk; it explains the label generally and gives Codex network access as an example of a capability carrying elevated risk of prompt injection attacks. Status: Confirmed Entry: tool/chatgpt/permissions-and-security/elevated-risk-features Revision: 1 Digest: sha256:048f3407dc080113 (changes whenever the content changes) Timeline: Created: 2026-11-13T03:10:48Z Last changed: 2026-11-13T03:10:48Z Last verified: 2026-11-20T11:48:59Z Next check: 2026-11-21T11:48:59Z Check interval: Daily Expected change: No fixed schedule Data: Field | Value | Source | As of example labeled feature | Codex | [1] | 2026-11-20 example capability | network access | [1] | 2026-11-20 common risk | prompt injection attacks | [1] | 2026-11-20 label icon | the exclamation mark within a shield | [1] | 2026-11-20 Sources: [1] OpenAI. "Elevated Risk labels". https://help.openai.com/en/articles/20001062-elevated-risk-labels Role: primary. Accessed: 2026-11-20T11:48:59Z Retrieved via: DeepSeek Fetch API (Markdown conversion; the original server response was not captured) Quote: "in Codex, our coding assistant, developers can grant Codex network access" (Supports: example labeled feature) Quote: "developers can grant Codex network access so it can take actions on the web like looking up documentation" (Supports: example capability) Quote: "The most common elevated risk that such labeled features have is prompt injection attacks." (Supports: common risk) Quote: "the exclamation mark within a shield" (Supports: label icon) Snapshot SHA-256: 9b71bb044ae774a32d80c203075e1efe92f2f33e6f9d3d18a60372ee6732f714 (captured 2026-11-13T03:10:47Z) 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]. [1] was retrieved through the DeepSeek Fetch API because a direct request did not return usable content. On this check the text of [1] was unchanged since 2026-11-13T03:10:48Z, so the values verified then were kept. Cross-check: not-available at 2026-11-20T11:48:59Z. No independent second source is configured for this entry Last successful check: 2026-11-20T11:48:59Z Last attempt: 2026-11-20T11:48:59Z Notes and limitations: Extracted automatically from a single official source. Related: https://latestn.com/tool/chatgpt/permissions-and-security/elevated-risk-label-meaning What does the Elevated Risk label mean in ChatGPT? https://latestn.com/tool/chatgpt/permissions-and-security/enable-auto-review How do I enable Auto-review in the ChatGPT desktop app? https://latestn.com/tool/chatgpt/permissions-and-security/enable-mfa How do I enable multi-factor authentication (MFA) on my OpenAI account? https://latestn.com/tool/chatgpt/permissions-and-security/dot-approval-link What happens when my dot needs approval for a request? https://latestn.com/tool/chatgpt/permissions-and-security/dot-camera-microphone-access Can my ChatGPT dot access my camera, microphone, or screen? https://latestn.com/tool/chatgpt/permissions-and-security/dot-password-handling How does my ChatGPT dot handle passwords? https://latestn.com/tool/ All entries in AI tools Change log: Time | Event | Change | Source 2026-11-13T03:10:48Z | added | Entry created | — Changes for this entry as a feed: https://latestn.com/tool/chatgpt/permissions-and-security/elevated-risk-features/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/chatgpt/permissions-and-security/elevated-risk-features POST: https://latestn.com/dispute/tool/chatgpt/permissions-and-security/elevated-risk-features with claim, evidence_url, and optionally field and contact (form-encoded or JSON) GET: https://latestn.com/dispute/tool/chatgpt/permissions-and-security/elevated-risk-features?claim={what is wrong}&evidence_url={link} How to cite: LatestN. (2026). Which ChatGPT features are labeled Elevated Risk? Revision 1, last verified 2026-11-20T11:48:59Z. https://latestn.com/tool/chatgpt/permissions-and-security/elevated-risk-features (accessed 2026-11-21T10:06:35Z). @misc{LatestN_tool_chatgpt_permissions_and_security_elevated_risk_features_r1, title = {Which ChatGPT features are labeled Elevated Risk?}, author = {{LatestN}}, year = {2026}, howpublished = {\url{https://latestn.com/tool/chatgpt/permissions-and-security/elevated-risk-features}}, note = {Revision 1; digest sha256:048f3407dc080113; last verified 2026-11-20T11:48:59Z; accessed 2026-11-21T10:06:35Z} } 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/chatgpt/permissions-and-security/elevated-risk-features