Recent searches
No recent searches
![Pam Couch's Avatar](https://secure.gravatar.com/avatar/4d12ede87455f3d1fd85fe9322e756e3?default=https%3A%2F%2Fassets.zendesk.com%2Fhc%2Fassets%2Fdefault_avatar.png&r=g)
Pam Couch
Joined Jan 24, 2023
·
Last activity Apr 17, 2024
Following
0
Follower
1
Total activity
22
Votes
0
Subscriptions
19
ACTIVITY OVERVIEW
BADGES
ARTICLES
POSTS
COMMUNITY COMMENTS
ARTICLE COMMENTS
ACTIVITY OVERVIEW
Latest activity by Pam Couch
Pam Couch commented,
December 12, 2023 - This Policy has been updated. For reference, a previous version of the Policy can be found above.
View comment · Edited Dec 12, 2023 · Pam Couch
0
Followers
0
Votes
0
Comments
Pam Couch created an article,
Across the Zendesk Suite, there are multiple features that are powered by generative artificial intelligence (AI). These features leverage large language models (LLMs) developed by third-party vendors that integrate with Zendesk.
This article lists the generative AI features available throughout the Suite, including links to more information about each feature as well as the associated LLM vendor.
This article contains the following topics:
Related articles:
List of generative AI features in Zendesk
The table below summarizes the generative AI features offered throughout the Zendesk Suite.
Feature | Description | LLM vendor |
Auto assist |
Understands the contents of submitted tickets and makes suggestions to your agents on how to solve them. |
OpenAI |
Suggested first replies |
Suggests a first response for agents in tickets based on existing macros and help center articles. |
OpenAI |
Custom intents (EAP) |
Classify submitted tickets using intents specific to your business needs. Personalizing intelligent triage by creating custom intents (EAP) |
OpenAI |
Ticket summaries |
Recaps all public comments that have been added to the ticket so far. |
OpenAI |
Expand ticket comment |
Adds additional language to the content of an agent’s comment. |
OpenAI |
Tone shift for ticket comment |
Changes the tone of an agent’s comment so that it sounds more friendly or formal. |
OpenAI |
Article summaries (EAP) |
Displays a concise description of a help center article at the top of the page. Using the new article editor with generative AI article summaries (EAP) |
OpenAI |
Quick answers in Agent Workspace |
Provides answers to searches within the Knowledge section of the context panel. |
OpenAI |
Similar tickets (EAP) |
Lets agents see a list of similar tickets to the ticket they’re currently working on. |
OpenAI |
Generative search for help center (EAP) |
Provides answers to search queries within your help center. |
OpenAI |
Expand for help center article |
Adds additional language to selected help center content. |
OpenAI |
Simplify for help center article |
Makes selected help center content clearer and more concise. |
OpenAI |
Tone shift for help center article |
Changes the tone of selected help center content so that it sounds more friendly or formal. |
OpenAI |
Generative replies |
Delivers automated answers in response to customer requests using information from your help center. |
OpenAI |
Bot personas |
Determines the style of expression applied to AI-generated messages to better reflect your brand. Using bot personas to add personality to AI-generated responses |
OpenAI |
Call summaries and transcriptions |
Generates a concise summary and full transcript on the ticket for each call. (Talk) Using generative AI to create call summaries and transcripts on tickets (Zendesk QA) Using Voice QA |
OpenAI |
About LLM vendors used by Zendesk
An LLM is a type of generative AI model designed to generate human-like language based on statistical patterns learned from vast amounts of text data. Zendesk integrates LLMs into many of our features to help you deliver exceptional customer service at scale.
Currently, Zendesk integrates LLMs from the following third parties into generative AI features:
LLM vendor | Data locality | Usage requirements |
OpenAI | Zendesk leverages OpenAI’s ‘zero data retention’ endpoints, meaning request and response bodies are not persisted to any logging mechanism and exist only in memory in order to serve the request. No data is stored by OpenAI following the delivery of the output. | Use of Zendesk services powered by OpenAI is subject to compliance with the OpenAI usage policies: https://openai.com/policies/usage-policies |
Edited Jan 27, 2025 · Pam Couch
3
Followers
30
Votes
0
Comments
Pam Couch created an article,
We have updated this Policy. If you are a new Subscriber, then this Policy will be effective as of December 13, 2023. If you are an existing Subscriber, we are providing you with prior notice of these changes which will be effective as of January 12, 2024. For the previous version of the Policy, please expand the section below.
Zendesk is committed to conducting business worldwide ethically and according to high standards, which includes taking steps to ensure our services are not used to promulgate abusive, harmful, or otherwise objectionable content.
The Policy is incorporated into the Zendesk Main Services Agreement (“MSA”) and applies to all Subscribers, Agents and End-Users who use the Services. Notwithstanding the terms of the MSA, Zendesk may, in its sole discretion, suspend or terminate an account (or any part thereof) and access to or use of the Services, and block future access to and use of the Services immediately due to a violation of this Policy. Zendesk may also remove or require a Subscriber to remove content published through the Services which Zendesk considers is illegal or which violates this Policy. Zendesk may take any other action such as disabling, restricting, or limiting access to, or distribution of, any such content as it considers appropriate.
We encourage all persons to report suspected illegal content and/or violations of the Policy by using the form above. There may be instances where we are limited in how we can respond to reported violations, but we always want to know about this activity. If you have a copyright infringement complaint, please review and follow the process outlined in Zendesk’s Copyright Infringement Notice and Takedown Policy. Zendesk may modify these policies at any time. Capitalized terms used in this Policy but not defined herein have the meaning given to them in the Zendesk MSA.
Subscribers are responsible for ensuring their Agents and End-Users comply with this Policy, including monitoring and removing, as necessary, content published through the Services in violation of this Policy.
Prohibited Content and Conduct
1. Spam or Unwanted Contact
The Services may not be used to send or host spam, including the sending of unwanted or unauthorized commercial content, unwanted or mass solicitation, or other forms of duplicative or unsolicited messages.
2. Illegal and Fraudulent Content or Activities
The Services may not be used to engage in illegal, or fraudulent activities or to directly or indirectly promote such activities, or to publish illegal or fraudulent content. Please note that Zendesk is limited in its ability to review fraudulent activity that may occur outside of the Services or to assess the legality of certain activities and may require that persons work with law enforcement to validate the fraudulent nature or the illegality of the activity before we are able to take action related to a reported Account.
3. Malicious Products and Activities
The Services may not be used to transmit viruses, malware, or other harmful code or content. This includes sending automated content or traffic, generated by bots or otherwise, that harms networks, services or other infrastructure of Zendesk or any third party.
4. Hateful Activities
The Services may not be used to transmit or host hate speech, advocate violence or terrorism, distribute graphically violent or obscene content, or to support or normalize hateful, racist, libelous, or discriminatory activities. Further, use of the Services by organizations that promote or normalize hate against individuals, hateful speech, violence, terrorism or whose primary purpose is inciting hatred or violence, is prohibited. To effectively comply with this policy, organizations whose primary purpose is to host, promote or otherwise contain user-generated content (“UGC”) must implement reasonably effective moderation policies and remedial procedures related to UGC that would violate these prohibitions.
5. Child Exploitation and Human Trafficking
Use of the Services for any content or activity that promotes or displays child sexual exploitation or abuse is prohibited. We will remove such content from the Service and take appropriate action. Further, uploading or sharing content that promotes, enables or discusses human trafficking is prohibited.
6. Harassment and Bullying
The Services may not be used to engage in abusive, harassing, or bullying behavior, including when contacting Zendesk for customer support. This prohibition also includes a prohibition on illegal dissemination of private information or media using the Services.
7. Deceptive Behavior and Account Hijacking
The Services may not be used to deceive individuals or impersonate third parties. Accessing another Account without the permission of the Subscriber that owns the Account is prohibited.
8. Intellectual Property
You may not use the Services in a manner that infringes a third party’s rights, including intellectual property rights. Additionally, if a primary purpose or result of your business is to infringe third parties’ intellectual property rights, then Zendesk reserves the right to suspend or terminate the Services, even if you do not use the Services directly for such infringement. Please note that Zendesk may be limited in its ability to assess alleged infringement that occurs outside of the Services, and may not be able to act on every report of alleged infringement.
Who is responsible for this Policy?
Zendesk will investigate all reports submitted in good faith, and take any and all action it deems necessary, including removing or requiring Subscribers to remove illegal or otherwise violative content, and canceling or suspending Subscriber accounts.
In addition, Zendesk employs a machine learning algorithm which identifies phishing ticket creation activity, and automatically suspends any new accounts identified associated with this activity.
Any suspected fraudulent, abusive or illegal activity or content may be referred to appropriate law enforcement authorities. Where required by applicable law, we will advise Subscribers and End-Users if we restrict or remove any content submitted to the Services or suspend or terminate access to the Services, and will provide a statement of reasons as to why such action has been taken.
Zendesk Internal Complaint Handling System
Subscribers impacted by Zendesk enforcement activities (other than those identified as sending deceptive high-volume commercial content) may request a review of the circumstances by following the procedures set forth in the notice sent by Zendesk to such affected Subscriber.
Edited Oct 10, 2024 · Pam Couch
8
Followers
282
Votes
4
Comments