> For the complete documentation index, see [llms.txt](https://docs.ggwp.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.ggwp.com/release-notes.md).
# Release notes
This is where we post updates to let you know about new features and improvements.
## July 29, 2026
### What's new
**New dashboard user role: Viewer**
* You can now assign a **Viewer** role to give team members read-only access to the dashboard. Viewers can see all dashboard data but can't take moderation actions. Ideal for stakeholders who need visibility without edit access. Admins can assign this role from the **Manage Users** page.
**Severity and language on custom blocklists** **(admin only)**
* Custom blocklist entries now support **severity levels** and **language**, giving you more precise control over how flagged terms are categorized and handled.
**Updated incident category taxonomy**
* You may notice some categories now labeled "legacy." We've preserved your existing categories as-is so nothing changes in your current setup or reporting while we roll out an expanded category set with broader content and language coverage and finer severity levels. Anything marked "legacy" continues to work exactly as it does today, so no action is needed on your end.
**More context on the Chat incident details page**
* The incident details page for chat now surfaces additional **incident metadata** (if applicable), giving moderators more context at a glance while reviewing.
**Filipino (Tagalog) language support**
## July 15, 2026
### What's new
**New severity levels: Very High and Very Low**
* Added two new severity levels: Very Low and Very High across the dashboard, including filters and displays, to provide more flexibility when tuning moderation policies and giving you finer-grained control over how content is classified and prioritized.
* If you have any saved filters, make sure you update them to include these new severity levels as appropriate (for example, if you have a filter for high severity incidents, you may want to update it to include both "High" and "Very High" severities.
* Note: This is applicable to [V3 pipeline only](/api-docs-chat/standard-package/chat-api-v3.md)
**Auto-assignment now respects moderator availability**
* When enabled, the auto-assignment engine skips moderators who are offline, so incidents are routed only to moderators who are online and available. You can turn this on per rule (it's off by default).
* Note: dashboard users must manually sign out to move to offline status by using the "Sign Out" option in the upper right corner of the dashboard. If you're planning to use this feature, make sure your moderator teams know that they'll need to manually sign out when their shift ends.
**Stronger login security with IP whitelisting**
* For users who authenticate outside of SSO-enforced domains, you can now restrict email/password login to approved (IP-whitelisted) network locations (for example, a trusted corporate VPN). This adds a network-level security layer on top of 2FA: even with valid credentials and a second factor, sign-in only succeeds from an approved location, reducing the risk of unauthorized access from unknown networks.
* Users on SSO-enforced domains continue to sign in via SSO as before.
* Setup is handled by the GGWP team — contact your account representative to enable it.
**UX improvement: Deselectable quick filters**
* Filters now support deselection — click a selected filter again to clear it, without resetting the whole view.
***
## July 1, 2026
### Updates
* **v3 chat pipeline updates -** our v3 chat endpoint now supports:
* Per tenant (game/app) model thresholds
* Per tenant and channel category sensitivity and risk profiles
* Note: Front end support for self-serve configuration is coming soon. In the meantime, please reach out to your GGWP rep for configuration support.
***
## June 17, 2026
### Dashboard updates
#### Metrics and reporting updates
The dashboard now includes expanded Community Health metrics, giving a consolidated view of community toxicity trends and content incident data across all channels. This release covers the Community Health tab; the Moderation Activity tab is in active development and will follow in an upcoming release.
**Users section**
The Users section surfaces player reputation and behavior trends over time.
* Active users by reputation — A stacked bar chart showing the distribution of active players across five reputation tiers (Very High, High, Medium, Low, Very Low) for the selected time period. Hovering over any bar shows a per-tier breakdown with both counts and percentages.
* Active users by ladder level — A bar chart showing the distribution of active players across ladder levels, for communities using a sanction ladder. Hovering over a bar shows the user count and percentage change from the previous period.
* Users with high severity incidents — A line chart tracking volume over time of users with at least one high severity incident, with totals on the left y-axis.
* User alerts (only applicable for customers with user alerts enabled) — A line chart tracking two alert categories over time: new grooming alerts and new mental health risk alerts. Both metrics include totals on the left y-axis.
All Users charts support three time windows:
* 7 days — daily data points
* 30 days — 3-day increments
* 90 days — weekly increments
**Content section**
The Content section surfaces incident volume and rate data, filterable by source channel.
* Source filter — Charts and metrics update based on the selected source: Chat, Voice, Discord, Reports, or Images. Only one source can be active at a time.
* Summary metrics — Three KPI cards at the top of the section display the primary volume and rate metrics for the selected source
* Incident rate over time — A line chart showing incident (or report) rate trends for the selected source and time window.
* Incident rate by severity — A multi-line chart breaking down incident rate by severity level.
* Incident rate by category — A donut chart showing incident distribution across content categories. Given the breadth of categories, the donut format is used to surface relative distribution at a glance.
The Content section uses the same 7-day / 30-day / 90-day time windows as the Users section, with equivalent granularity.
**Coming soon: Moderation Activity tab**
The Moderation Activity tab is currently in development and will be released in a subsequent update. It will include:
* User moderation — Warning and sanction volume broken down by automated vs. manual actions, reinstatements, and ladder escalation/de-escalation activity.
* Content moderation — Live filter volume, unreviewed assigned incidents by channel, review outcomes by source, appeals data, and time-to-action metrics.
**Notes**
* All data shown in the dashboard reflects activity within the selected time window for your configured game title(s).
* Export functionality is available via the Export button in the top right. The exportable report remains unchanged in this release. Improvements to the exportable report are planned for a future release.
#### Image moderation support
We're excited to introduce image moderation support in the GGWP moderation dashboard. Image incidents now flow through the same queue, tooling, and workflows your team already uses for chat.
With image moderation integrated directly into your existing incident workflow, your team can respond to visual policy violations with the same speed, consistency, and audit trail you expect from chat moderation, while protecting the moderators doing the work.
**Viewing image incidents in the Incident List**
Flagged images now appear as incidents in the Incidents List page. Select “Image” from the source dropdown selector to see the list of image incidents.
Just like Chat and other sources (Voice, Usernames, Discord) each Image incident has a severity, confidence, incident type, and timestamp. The same moderation workflows are also supported, so you can assign Image incidents for review and mark review status like other incident types.
Filters work the same too, so you can filter the list of image incidents by criteria like incident type, severity, assignee, and more.
**Reviewing an image incident**
Click on any image incident ID to open the incident information modal. Opening an image incident shows the full incident context, including data source, severity, confidence, and incident type.
Built with moderator well-being in mind, flagged images are never displayed automatically. Click on “See flagged image” to open the image viewer. Images open blurred by default, and moderators have options to progressively reveal them, so they’re only exposed to harmful content to the extent needed to make a moderation decision. This deliberate-reveal interaction reduces unnecessary exposure to harmful content and keeps your team in control of what they see.
Image incidents support the same status workflow (e.g., Unreviewed, Escalated) and accuracy confirmation as chat incidents, so your QA and reporting stay consistent across content types. Previous/Next navigation lets reviewers move efficiently through the queue without returning to the incident list.
**Note**
* Image incidents will appear automatically in the incidents list for titles with image moderation enabled. Contact your GGWP account team to enable image moderation for your title or to learn more.
#### New filter options
There are two new quick filter options that admins can configure from the Content Filtering page:
* **Age Concern**
* Flags messages in which a user explicitly or implicitly indicates their age. This category is informational and used to enable age-aware safety controls rather than to signal harmful intent. The default setting for this category is OFF.
* Three sensitivity levels are supported:
* Low - Under 13
* Indications that a user is a child below the age of 13, often triggering the strictest safety and compliance rules.
* Examples: "I'm 10", "I'm 12 years old", "I'm in elementary school", "I'm a kid"
* Medium - Under 18
* Indications that a user is a minor but at least 13 years old. Used to apply enhanced protections and monitor interactions with adults.
* Examples: "I'm 15", "I'm 17", "I'm in high school", "I'm a minor"
* High - Under 21
* Indications that a user is a young adult below 21. Typically relevant for region-specific policies (e.g., alcohol, gambling, mature content).
* Examples: "I'm 19", "I'm not 21 yet", "I can't legally drink", "I'm 20"
* **Minor Safety**
* Flags language that may place minors at risk by indicating grooming behavior, predatory intent, sexualization of minors, or attempts to exploit, manipulate, or endanger children, as well as references to exploitation or sexual abuse of children (excluding first-person disclosures or statements from potential victims). This category captures risk to minors, regardless of whether a minor is present in the conversation.
* Three sensitivity levels are supported:
* Low (default setting) - Sexual Exploitation, CSAM, or Explicit Sexual Targeting of Minors
* Sexual content involving minors where the speaker is directly targeting a minor or expressing unambiguous predatoryintent toward minors. This includes first-person expressions of sexual attraction to minors, endorsement of sexual acts with minors, requests for sexual images from minors, possession or promotion of child sexual abuse material (CSAM), or direct sexual communication toward a known minor. This includes statements that actively normalize or justify sexual activity with children
* Examples: "I like little kids sexually", "kids should be taught sex early", "I want to fuck minors", "children are better partners", "I enjoy abusing kids", "I have 10 TB of cp", "can’t wait to see underage girls"
* Medium - Grooming Signals or Inappropriate Adult-Minor Interaction
* Flags language suggesting grooming behavior, inappropriate emotional trust-building, secrecy, or attempts to isolate, privately engage with, or build inappropriate relationships with a presumed minor.
* Examples: "you're mature for your age", "don't tell your parents we're talking", "you can trust me more than your parents", "let’s chat somewhere adults won’t see", "you seem older than other kids your age"
* High - Boundary Testing and Minor Safety Concerns/References
* Flags language that raises potential minor safety concerns through personal questions, secrecy, or boundary-testing behavior without clear sexual intent or explicit grooming, as well as any reference to exploitation or sexual abuse or children (excluding first-person disclosures or statements from potential victims). These messages may indicate attempts to build familiarity, assess vulnerability, or initiate more private interaction, or they may represent sensitive discussions surrounding minor safety.
* Examples: "do you have a boyfriend?", "we can keep this between us", "what school do you go to?", "are you home alone?", "how old are you really?", "you're a pedophile", "he was arrested for pedophilia", "i can't image being targeted for grooming"
#### Other updates
* Automod - support for rules based on content category
* In addition to configuring automod rules based on criteria like incident confidence, severity, user reputation, and recent incident history, you can now also configure specific rules based on content categories (e.g. Identity Hate, Verbal Abuse, etc).
* Please reach out to your GGWP representative to discuss automod configuration
* Improvements to live filtering precision across Chinese, Japanese, and Vietnamese
* Expanded username link sharing detections to cover a broader range of violations
***
## Apr 22, 2026
### Dashboard updates
* **Grooming alerts update** - ability to opt-in to email notifications for grooming alerts
* Users now have the option to opt in to grooming alert email notifications
* To opt in, go to the user avatar icon in the upper right corner of the dashboard and click to open the dropdown menu and use the toggle to opt in or out of email notifications
* When enabled, you'll receive email notifications for new alerts and updates to existing alerts
* **Incident information modal - added `live filtered` flag for live filtered chat messages**
* For chat messages, the incident information modal (accessible via the incident list) now shows if a message was flagged for live filtering
* You can use the toggle next to the message to switch between original and filtered message views, with the filtered version as the default view
* **SCIM support updates**
* Dashboard administrator roles are now restricted from managing users when SCIM is enabled
#### Other updates
* **Hindi and Persian language support**
* **Image moderation for Discord**
* The Discord bot now supports image moderation using the GGWP Image Moderation API
* Please contact your GGWP rep if you'd like this enabled
***
## Apr 8, 2026
### Dashboard updates
* **New feature - SCIM support**
* SCIM (System for Cross-domain Identity Management) 2.0 API enables standardized user provisioning and lifecycle management across systems. This set of APIs allows identity providers (IdPs) such as Okta, Azure AD, or others to:
* Create users
* Update user attributes
* Delete users
* Retrieve user information
* Documentation is [here](https://docs.ggwp.com/identity-and-access-management-iam/api-docs-scim/introduction). Please reach out to your GGWP account rep if you have any questions.
* **New feature - Incident auto-assignment**
* Admins can now auto-assign new incidents to specific moderators based on language.
* Multiple moderators may be selected per language, and new incidents will be distributed evenly across those moderators.
* **Custom allowlists and blocklists now support regex**
* Regex enables powerful pattern matching but can cause errors or performance issues if written incorrectly. Please use this feature only if you're familiar with regex syntax.
* **Player reports update**
* For scenarios where a user can report a specific message or post sent by another user, the Player report insert API now supports providing message\_id (optional) in order to link the report with the reported message.
* The Incident Details page in the dashboard will now show both the reported message (if applicable) as well as any other recently detected incidents.
* Incidents list - the "Prioritized" toggle now defaults to "off"
### Other updates
* **Image moderation API**
* You can use this to process image data and integrate moderation actions directly into your application workflow. This product is designed to detect and classify a wide range of harmful or policy-violating content present in images.
* Documentation is [here](https://docs.ggwp.com/api-docs-image). Please contact your GGWP account rep to get started with image moderation.
* **Dutch language support**
***
## Feb 25, 2026
### Dashboard updates
**Content Filtering option update for PII**
* The Personally Identifiable Information (PII) category for chat and Discord now includes medium and high filter strength options.
* The medium setting filters out content that strongly appears to contain PII based on recognized patterns and context.
* The high setting filters out all content that matches common PII formats, regardless of confidence level.
* If your configuration is currently set to "on" for PII, your setting will now default to medium. If your PII setting is set to "off," it will remain off. Dashboard admin users can change these settings anytime from the Filter Settings page (accessible via the Content Filtering option in the left nav).
**Incident detail page updates**
* Incident details for Player Reports now contain the session conversation for added context surrounding the incident, if applicable.
* The "Reported User" section is now expandable and collapsible.
### Other updates
* **Finnish language support**
***
## Dec 18, 2025
### Dashboard updates
**SSO and MFA support**
* We now support SSO and MFA login options at the organization level. Please contact your GGWP rep for details on how to get onboarded.
**Player detail page**
* You can now save your filter settings for the incident list at the bottom of the player details page.
* Your saved filter will persist across all player detail pages and can be updated at any time.
**Incidents list - saved filters**
* You can now create up to five saved filters in the incidents list.
* How saved filters work:
* In the left nav, you can access all your saved filters under the Incidents tab. You can expand or collapse this view using the caret icon in the tab.
* To save a filter, select the "Save Filters" option in the top right of the incidents list.
* You can choose to save a new filter or overwrite an existing filter.
* From the left nav, you can also rename or delete saved filters using the 3-dot menu to the right of each filter.
* The filters you save only appear for you, so other users within your organization won't see them.
***
## Nov 19, 2025
### Dashboard updates
**Updated login flow**
* There are now two steps to log in to the dashboard. On the first screen, you'll enter your organization name, and on the next screen, you'll enter your email and password.
**Moderator notes**
* You can now add an optional note when you manually sanction or reinstate a player to give context to other members of your team about why you took the action.
* You can view notes in the sanction history within the player details page. Team members can view each other’s notes, and the person who originally posted the note has the option to edit or delete the note.
**New quick filters**
* There are two new quick filter options that admins can configure from the Content Filtering page:
* Scam
* Flag messages that contain fraudulent schemes aiming to steal accounts, in-game/real-world assets, or high-risk personal information (e.g., social security number, credit card information, physical address) through deceptive, manipulative, or dishonest tactics.
* Examples: “You’re selected to claim the sum of $100,000 in the lottery”, “please send me ur login access”, “F A K E, COM Sell cheap diamonds! Code:P5=5% OFF!.…”
* Solicitation
* Flag messages that attempt to obtain another player’s low-risk personal information (e.g., name, phone number), lure them off-platform, or extract real-world value (e.g., money, digital assets) through direct or suspicious requests.
* Examples: “please send me 20 dollars”, “We can chat on Instagram if you want”, “Can i have your phone number”
* These categories are set to “off” by default, so you’ll need to turn them on in the dashboard to enable them.
**Players list - new filter option**
* You can now filter by sanction ladder level in the player list
**Incident information modal updates**
* We’ve made some small copy updates to the incident information modal to use clearer and more accurate language.
* The “Incident accuracy” dropdown section is now called “Incident feedback”
* The “False Positive” option in the incident feedback dropdown is now called “Dismiss” to be inclusive of scenarios when the incident is accurately flagged but you want to dismiss it because the content is permissible within your community’s policy.
* Note that there are no functionality changes to these dropdown menu options - everything works the same as before, and only copy has been updated to better reflect how this feature is used.
***
## Oct 9, 2024
### Dashboard updates
* **Improvements to automoderation support for player reports**, including adding the sanction ladder for reports on each player detail page
* **Player list page updates**
* "Sanction status" column now supports showing multiple statuses if a player has an active sanction across more than one channel
* Improvements to filtering options, including ability to filter auto-sanctioned players by channel and more flexibility when choosing date ranges
### Other updates
* Reputation Batch API now allows you to fetch the latest reputation scores for all users in CSV format. You can find more details [here](/api-docs-reputation/reputation-score-batch.md)
* Webhook version 1.1 simplifies the payload validation of webhook calls
* Chat API updated to restrict the input message length to 1,000 characters to ensure maximum performance
***
## July 17, 2024
### Dashboard updates
**New feature - Custom allowlists and blocklists**
* Custom allowlists and blocklists allow you to create custom overrides to GGWP's global default blocklists.
* For example, if there is a term that's on our blocklist that you want to allow, you can add it to your custom allowlist to override global defaults.
* If there's a term that's not included in our default blocklist that you would like you block, you can add it to your custom blocklist.
* Also note that common variants for terms are automatically captured by our system, so there's no need to add all variants of a term (e.g., if you add "sandwich" to your blocklist, additional terms like "sandwiches", "sandw1ch", etc will also be blocked).
* Note: you must have the `admin` role in order to access this feature.
* To view your custom allowlists and blocklists, click the gear icon in the upper right of the dashboard and select `Allowlists and Blocklists`.
* On the custom allowlists and blocklists page, you'll see tabs for both `Chat` and `Usernames`, and you can click on the tabs to navigate to the lists you want to edit.
* From here you can search for terms, add terms, or delete terms.
* Make sure you save and publish your changes when you're done making updates.
* Note that if your organization includes multiple games, custom allowlists and blocklists are at the game level, so if you want to make changes across multiple games you'll need to update each list separately.
***
## June 19, 2024
### Dashboard updates
**New feature - Grooming alerts**
* This feature flags players who may be engaging in grooming behaviors so that your team can review each player’s activity and take action as needed.
* You can access grooming alerts by clicking on the `Alerts` option in the left menu. On the alerts page, you’ll see a list of players sorted by risk level.
* From there you can view more details on each player to view suspicious messages, including the ability to:
* See each message in context of the chat by clicking the icon to the right of the message.
* “Pin” specific messages (using the pin icon to the left of the message) and filter by pinned messages, in case there’s a player you’re reviewing and want to keep track of specific messages they’ve sent.
* Export a user's chat history and download as a csv, with the option to export their entire chat history or only chat sessions that contain risk indicators
* There are also different statuses you can set for each alert, like “watching” for players you want to keep an eye on, “confirmed” for confirmed grooming cases, and “dismissed” for cases that don’t look suspicious. Note in the case of “dismissed” alerts, if the player engages in new suspicious activity then that will trigger a new alert, so dismissing won’t permanently disable alerts for that player.
* Player list page
* Added a new `Auto-sanctioned` filter toggle to show only auto-sanctioned players when toggled on
* Added two new dynamic columns with data for `Sanctioned On` and `Sanctioned For` when `Auto-sanctioned` filter is active