This Artificial Intelligence (AI) Policy and Notifications document (this “AI Policy”) applies to all Services offered by Company and contains important disclosures, notifications and disclaimers to which the entity purchasing any of the Services (“You”) and Your Users will be subject. This AI Policy is intended to provide You and Your Users with additional information regarding Company’s AI functionalities within the Services, including guidelines for the use of such functionalities and the handling of Your personal data and other confidential or sensitive data within the AI functionalities.

This AI Policy is issued pursuant to and incorporates by reference the terms and conditions of the Master Service Agreement (the "MSA") by and between Company and You and serves as a supplement to the Company Product Schedules published with respect to the Services. All capitalized terms in this AI Policy shall have the same meaning as set forth in the MSA, unless defined herein.

By ordering the Services, selecting “I Accept and Continue” in Your Account, agreeing to the terms of the MSA, or using the Services, You acknowledge and accept this AI Policy and agree to be bound by its terms. This AI Policy is effective immediately upon Your initial use, order or purchase of the Services on or after the effective date hereof.

In the event of a conflict or inconsistency between the terms of the MSA and the terms of this AI Policy, this AI Policy shall supersede and govern.

Definitions. For the purposes of this AI Policy, the following definitions apply:

1. Background; Use of Third-Party Solutions.

  1. AI features offer the capability to perform numerous tasks more efficiently and accurately, such as analyzing data, creating content, making predictions and automating tasks. Because AI has such tremendous potential to enable greater efficiency and expanded insights into data and communications, Company intends to continue developing and introducing new AI features and functionality to its Services to help You realize those benefits. However, AI is an extremely complex and rapidly evolving area that presents numerous potential risks, limitations and restrictions related to issues such as confidentiality, privacy, intellectual property and reliability, to name just a few. The purpose of this AI Policy is to help You better understand those risks, limitations and restrictions so that You and Your Users can make informed decisions regarding Your use of Company’s AI features and functionality, including whether to use such features and functionality, how to use them and their outputs, who should have access to them, and what types of data to share with them. By activating and using Company’s AI features and functionality, You (a) agree that You and Your Users shall use the AI features in a secure, responsible and confidential manner, in accordance with applicable law and regulations; (b) acknowledge that You have been advised of and accept the risks, limitations and restrictions associated with using AI, including without limitation those described in this AI Policy; and (c) agree that it is Your responsibility to advise Your Users regarding the risks, limitations and restrictions related to AI features described herein.
  2. Many of the AI-related services, features and functionality offered by Company are powered by third-party AI solutions. Company attempts to identify and use leading third-party AI solutions, and Company conducts thorough reviews of the security (of both the third-party AI tools themselves and the infrastructure used to host and deliver them) and performance of such solutions prior to selecting the third-party vendors to provide such services. In addition, Company will use reasonable care to securely transmit to the third-party AI service providers the data that Your Users submit to the AI tools. However, Company has no control over the quality or performance of such third-party AI solutions. Furthermore, the data that Your Users submit to the AI tools (along with the output generated by the applicable AI tool) will be stored, transmitted and processed by such third-party service providers (including without limitation providers of the AI solutions themselves and providers of data storage for such information), and Company has no involvement in the processes or practices that such third-party vendors use when handling Your and Your Users’ data. Company therefore makes no assurances regarding, and disclaims all liability and responsibility regarding, (a) the quality, accuracy and/or performance of any such third-party AI solutions; and/or (b) the ability of the providers of such third-party AI solutions to adequately and successfully protect the privacy and/or confidentiality of any of Your data that You or Your Users provide to or receive from such AI solutions.

2. General Limitations of AI.

  1. When using AI, You and Your Users should take into consideration a variety of disclaimers and limitations to help enhance the safety, accuracy and effectiveness of the output of AI technology. This is particularly relevant when using and relying upon content generated by Generative AI features. Examples include:
    1. Potential for Errors/Inaccuracy: AI systems are not always accurate and error-free. AI systems can make mistakes when processing and/or generating information, or the information they are accessing may not be correct. For example:
      1. The source data that is accessed by an AI system is generally a fixed set of data as of a particular date. That source data may become outdated, and an AI system will generate incorrect output if it relies on any such outdated data when generating that output.
      2. The source data that is accessed by an AI system may be incorrect. AI systems generally refer to data available on the Internet, some of which may be factually inaccurate. Reliance on any such inaccurate data will lead to errors in the output generated in reliance on that data.
      3. Even if source data is current and accurate, AI systems may not process or interpret that data correctly, and the resulting output may therefore be misleading or incorrect.
    2. Security Risks: Like any computer system, AI systems can potentially be vulnerable to cyber attacks. Users should be cautious when using AI tools and should avoid sharing sensitive information (such as confidential data or personally identifiable information) with AI systems, whenever possible. Examples of potentially confidential or sensitive information may include:
      • Proprietary source code or product diagrams;
      • Passwords;
      • Lists or documents containing customer or vendor information;
      • Financial data or pricing information;
      • Personally identifiable information (or PII), such as individuals’ names, social security numbers, addresses, email addresses, phone numbers, or any other data that can uniquely identify an individual;
      • Credit card or bank account information;
      • Employee data;
      • Copies of business contracts; and
      • Security information
    3. Bias Issues: AI systems are trained on particular sets of data. That training data may be biased. As a result, the output from AI systems can be biased due to issues with the data on which they are trained. Users of AI should be aware of potential biases and take steps to mitigate them, such as by independently verifying statements or findings generated by AI tools.
    4. Legal Compliance: The legal and regulatory environment related to AI tools is constantly and rapidly changing. Governments around the world are considering legislation that would restrict or limit the ways that AI systems can function and can be used. In addition, the output of AI systems may potentially violate the intellectual property or other rights of third parties, such as by not properly identifying information that is owned (and whose use is restricted) by others and incorporating that information into its output. Users must always be aware of these risks and the importance of using AI systems and tools (and their output) in a legally compliant manner. When using a Generative AI tool to generate computer code, for example, it is recommended to perform a code scan on the generated code to identify security vulnerabilities and/or potentially problematic open source/third-party code before incorporating such code into a company’s products.
    5. Ethical Considerations: AI systems may raise ethical concerns, such as potentially violating privacy laws or regulations or reflecting discriminatory or offensive content. Users should ensure that all output of an AI tool considers and evaluates the sensitivity and appropriateness of the underlying data and does not use or embody data in an unethical manner.
  2. In general, users should approach AI with caution and be aware of its limitations and potential risks. It is crucial that users use AI responsibly, carefully review the output of any AI tool before using it, and take steps to ensure that such tools are used safely and appropriately.

  3. Certain third-party AI tools are structured to use the content of customer communications to train their tools. However, Company has designed its AI-related services, and its use of third-party AI services in connection therewith, with the intent of not using any of Your audio, video, chat, screen sharing, attachments or other communications using Company’s Services, without Your consent, to train Company’s or third-party artificial intelligence models.

3. AI Notifications Regarding Specific Company Services and Features.

  1. AI Assistant. Company’s AI Assistant is a Generative AI-powered chatbot that enables users to converse with it through the chat function within Company’s Unified Communications service. All of the limitations and disclaimers associated with using AI apply to the AI Assistant. In addition:
    1. If You purchase Company’s Unified Communications Archiving (“UC Archiving”) service, You and Your Users should be aware that (a) the UC Archiving service is configured to archive all communications between Your Users and the AI Assistant (just as the service archives all of Your Users’ other chat communications), and (b) as a result, Your Account administrators will have the ability, by accessing the archive of Your company’s chat communications, to view Your Users’ interactions with the AI Assistant.
    2. The AI Assistant has contextual memory (i.e., the ability to access and reference previous communications between a User and the AI Assistant), but that memory is limited to a small number of previous communications (typically, the most recent three messages) and only for a limited period of time. If a User wishes to reset the AI Assistant’s contextual memory, the User can do so by entering “/newchat” into the chat box with the AI Assistant.
    3. Users are able to send URLs or other forms of web links to the AI Assistant, and the AI Assistant may include URLs or web links in its output. Company does not scan such URLs or web links for malicious code, inappropriate or illegal content, or any other harmful attributes, including without limitation disabling devices, drop dead devices, time bombs, trap doors, Trojan horses, worms, viruses and similar mechanisms (collectively, “Harmful Content”). Company expressly disclaims all liability with respect to any Harmful Content contained in any URL or web link shared by or with Your Users via the AI Assistant. You hereby release and agree to hold harmless Company from and against any damages or liabilities of any kind related to any Harmful Content contained in any URL or web link shared by or with Your Users via the AI Assistant.
    4. Company does not screen or filter the content of messages, links or attachments sent to or by the AI Assistant (whether for offensive or illegal content, viruses or otherwise), and Company does not modify any User content sent to the AI Assistant. As with all Services, You assume full responsibility and liability for the legal and compliant use of the AI Assistant by Your Users. Company expressly disclaims all liability with respect to any content, links or attachments included by Your Users or any third parties in a message to the AI Assistant or in any response or output generated by the AI Assistant.
  2. AI-Powered Call/Meeting Transcriptions. Company offers an AI-powered transcription feature for voicemails and recordings of phone calls and online meetings. All of the limitations and disclaimers associated with using AI apply to the AI Call/Meeting transcription features. In addition:
    1. Any data that is submitted to the Generative AI tool by any participant in a transcribed call or meeting may be included in the transcription of that call and therefore may be accessible to the Generative AI tool as referenceable data for generating future work product for that particular customer.
    2. You and Your Users should be aware that Your Account administrators will have the ability to view, and the ability to allow supervisors and coworkers to view, any transcriptions of phone calls or online meetings.
  3. AI Call Summarizations and AI-Powered Redactions. Company’s AI Call Summarization is a Generative AI-powered tool that generates a high-level summary of a User’s phone call over Company’s Contact Center service. Company also offers an AI-powered redaction function that seeks to identify and redact sensitive information, such as social security numbers or bank account data, from transcriptions of phone calls through Company’s Contact Center service, which are used to create the call summaries described earlier in this paragraph. All of the limitations and disclaimers associated with using AI apply to the AI Call Summarization and redaction features. In addition:
    1. While Company uses reasonable efforts to identify and extract/redact personal, sensitive or other confidential information from the call transcriptions, the AI-powered redaction tools may not successfully identify and redact all sensitive and/or confidential information that is included in a call transcription. As a result, sensitive and/or confidential information that is communicated during a call may potentially be included in transcriptions and summaries of that call and may therefore be visible to Account administrators, supervisors and coworkers that have access to such transcriptions and summaries. In addition, any such sensitive and/or confidential information, if it is not successfully redacted from a call transcription or summary, may be included in the database of information that is made available to the Generative AI tool for future use and reference by the particular customer. Any data that is submitted to the Generative AI tool by any participant in a summarized call or a transcribed call may potentially be included in a call summary or transcription and/or accessible to the Generative AI tool as referenceable data for generating future work product for that particular customer.
    2. You and Your Users should be aware that Your Account administrators will have the ability to view, and the ability to allow supervisors and coworkers to view, any summarizations of phone calls over Company’s Contact Center service.
  4. AI-Based Email Protection. Company’s AI-Based Email Protection is a security software offering that uses AI to identify potential email threats by analyzing typical communication styles, trends and terminology. All of the limitations and disclaimers associated with using AI apply to the AI-Based Email Protection offering. In addition:
    1. The AI-Based Email Protection offering is on, by default, for every Hosted Exchange mailbox. You can turn off the AI-Based Email Protection offering at any time in Company’s administrative control panel. The AI-Based Email Protection offering relies on machine learning to identify patterns and trends in each User’s communications. Company does not guarantee that the AI-Based Email Protection offering will successfully identify every malicious email or will correctly characterize each inbound email as safe, dangerous or otherwise, and Company expressly disclaims all liability with respect to the performance of the AI-Based Email Protection offering.
  5. AI Support Bot. Company’s AI Support Bot is an AI-powered bot that has access to Company’s support-related articles, instructions and frequently asked questions, enabling partners and end customers to more easily access self-service technical support information and obtain online support assistance. All of the limitations and disclaimers associated with using AI apply to the AI Support Bot. In addition:
    1. The AI Support Bot will be able to access multiple sources of Company technical support information to identify the desired answer to a User’s query. However, the data set to which the AI Support Bot has access, and that is used to train the AI Support Bot, must draw data from multiple different sources where the underlying content is posted. Because that data is dispersed in multiple locations (e.g., Knowledge Base articles will be stored in one location/system, and Frequently Asked Questions will be posted and stored in a separate location/system), instead of in one centralized location, the AI Support Bot may not be able to look to multiple different data repositories and successfully identify correct responses to a User’s query.
    2. Company will have access to review Users’ interactions with the AI Support Bot (both queries and responses) in order to evaluate and help improve the accuracy and performance of the AI Support Bot.

4. Service Limitations and Disclaimers.

  1. Service Limitations. The AI tools, features and functionality offered through or in connection with the Services are provided on an “as is” basis, and Company makes no representation or warranty with respect to the availability, the effectiveness, the accuracy or any other aspect of the output or performance of such AI tools, features and functionality. Use of the AI tools, features and functionality is at Your sole risk. Notwithstanding anything to the contrary otherwise set forth herein, You hereby release and agree to hold harmless Company from and against, and that Company will have no liability whatsoever in connection with, any damages or liabilities of any kind arising out of:
    1. Any inaccuracies, errors, biases or offensive content contained or reflected in any output generated by an AI tool, feature or functionality, regardless of whether such inaccuracies, errors, biases or offensive content are the result of incorrect, outdated or biased source information, failures of the AI tool to properly process source information and generate correct output, or any other reason;
    2. Any security incident impacting Your data that results from a failure by any third party provider of AI-related services to adequately protect such data from unauthorized access, use, misappropriation or other conduct;
    3. Output from any AI services, features or functionality that (i) violates any applicable law or regulation; (ii) infringes on the intellectual property or other rights of any party; (iii) is subject to restrictive licensing provisions or other restrictions or limitations that could negatively impact any party that attempts to use such output;
    4. Any failure by a third-party provider of AI-related services to successfully segregate Your data from the data of other users of the provider’s services; and
    5. Any failure of an AI tool to correctly identify and redact/remove sensitive or confidential information from a summary, transcript or any other work product that it generates.
  2. Company Service Level Agreements do not cover the AI features and services (for example, unavailability of an AI feature or service does not constitute unavailability of the applicable Service in which such AI feature or service is used, for purposes of the applicable Service Level Agreement of such Service). Company does not offer a separate Service Level Agreement for the AI features or services. 

    You acknowledge and agree that the limitation of Company’s liability is a material term to Company and that it would not otherwise make the AI tools, features and functionality available without this limitation, and that You agree these limitations are reasonable.

  3. Limitation of Liability. Company will not be liable for any direct, indirect, incidental, special, punitive or consequential damages, including but not limited to damages for lost profits, business interruption, loss of programs or information, and the like, that result from the use or inability to use the AI features and services or from mistakes, omissions, the AI features or services not meeting Your requirements or expectations, interruptions, errors, defects, or delays in operation or transmission, regardless of whether Company has been advised of such damages or their possibility. Company will not be liable for any harm that may be caused by the execution or transmission of malicious code or similar occurrences, including without limitation disabling devices, drop dead devices, time bombs, trap doors, Trojan horses, worms, viruses and similar mechanisms. The warranty restrictions and limitations set forth in this Schedule are in addition to the warranty restrictions and limitations provided for in the MSA. Except as otherwise set forth in this Schedule, You agree that the total liability of Company and Your sole remedy for any claims shall be as set forth in the MSA.

5. Fair Use Policy.

  1. Company may offer AI features or services that are described as including unlimited use. However, Company reserves the right to review usage of any “unlimited” plans to ensure Fair Use and reserves the right to, at any time, take the actions noted in Section 5.2 below without notice to You. "Fair Use" means that the total usage of the applicable Service (whether measured in terms of queries, volume of content transmitted to or processed by such Service, or any other metric that reasonably reflects the cost incurred by Company to deliver such Service) does not substantially exceed, in Company’s reasonable judgment, the average use of all other Company customers as measured on a per user basis. Usage and associated charges for excess usage will be determined based solely upon Company’s collected usage information. Fair Use also prohibits any activities that result in excessive usage including, but not limited to, automated queries, continuous or extensive recording of calls or meetings, mass uploads of data, or any activity that disrupts the activities of Company and/or other Company customers.
  2. If Your usage exceeds the limits for Your Account (if applicable) or otherwise exceeds Fair Use, You agree that Company may immediately, in its sole discretion, (a) charge You for such excess usage via Your automated payment account (or by invoice if You have been accepted into Company’s check paying program), (b) upgrade You to a plan or increase the limits on Your Account to address this excess usage, and/or (c) suspend Your Account or terminate Your Account upon notice to You. Upon any upgrade or increase on the limits of Your Account, You will be responsible for the new costs and fees.

6. Use of the AI Features and Services.

  1. Business Use. You will use the AI features and services predominantly and primarily for Your own internal business, non-personal use. You will not allow any third party, including Your vendors and service providers, to access or use the AI features and services.
  2. Restricted Activities. You will not use the AI features and services: (a) for activities that are unlawful, harassing, libelous, abusive, harassing, tortious, defamatory, threatening, harmful, invasive of privacy, vulgar, pornographic, obscene or otherwise objectionable in any way or that are harmful to minors in any way under the law or otherwise; (b) to transmit or knowingly to accept any material or communications that may infringe the intellectual property rights or other rights of third parties, including, but not limited to, trademark, copyright, patent or right of publicity; (c) to interfere with, disrupt, attempt to interfere with or disrupt computer servers or networks connected to the Services or violate the regulations, policies or procedures of such networks; (d) to harass or to interfere with another user’s use and enjoyment of the Services; (e) to transcribe or otherwise process unlawfully recorded conversations or meetings in violation of applicable law; or (f) in a manner deemed by Company to be inappropriate. You may not access the AI features and services for purposes of monitoring their performance, availability, or functionality, or for any other benchmarking or competitive purposes, without Company’s prior written consent. You may not access the AI features and services if You are a direct competitor of Company, without Company’s prior written consent pursuant to a separate written agreement.
  3. No Resale. The AI features and services are for Your use and not third parties. Except as set forth in the following sentence, You shall not resell the AI features and services or make them available for use by third parties, and doing so constitutes an abusive practice subjecting You to immediate termination of this Schedule and the Services. You acknowledge and agree that in order to resell the Services, You must enter into a separate written agreement with Company.
  4. Applicable Law; Fraudulent Usage.
    1. Applicable Law. You acknowledge and agree that access to and use of the AI features and services may be restricted or limited as a result of Applicable Laws and that You will not use, or allow the use of the AI features and services, in contravention of, and will comply with, any Applicable Law. You represent that (i) You and Your Users are not named on any Governmental Authority list of persons or entities prohibited from receiving exports, and (ii) You will not permit Users to access or use AI features or services in violation of any export embargo, prohibition or restriction. You acknowledge and agree that that it is Your sole responsibility to use the Service in a lawful manner. Notwithstanding any other provision of this Schedule or the MSA, this Schedule is subject to change, modification, or cancellation, with or without prior notice, for any reason, including without limitation as may be required or reasonably deemed necessary by Company pursuant to any Applicable Law, including any order, rule or decision of a Governmental Authority.
    2. Fraud. It is the express intention of the parties that You, and not Company, will bear the risk of loss arising from any unauthorized or fraudulent usage of AI features or services provided under this Schedule to You. Company reserves the right, but is not required, to take any and all action it deems appropriate to prevent or terminate any fraud or abuse in connection with the AI features or services, or any use thereof; provided, however, that any such action will be consistent with applicable federal, state and local laws, rules, and regulations and provided further that the failure to take any such action will not limit Your responsibility for all usage of the Services.
  5. Call and Meeting Recording. Certain of the AI features and services involve the recording of the content of phone calls or online meetings and/or the processing of data from such recorded meetings. Notwithstanding any other applicable provisions or prohibitions of use set forth in this Schedule or the MSA, You agree and acknowledge that You are solely responsible for complying with all applicable laws in any relevant jurisdiction when using call or meeting recording features. Company expressly disclaims all liability with respect to Your recording of conversations. You hereby release and agree to hold harmless Company from and against any damages or liabilities of any kind related to the recording of any conversations or meetings using, or the processing of any such recordings by, the AI features or services.
  6. Company’s Right to Remove or Modify AI Functionality. Company may, in its sole discretion, change or discontinue any AI feature or service at any time, without notice, and does not represent or warrant the result of any such action. Company may convert any AI feature or service to a paid service upon notice to you.

AI Policy and Notifications v. 2023.1
Effective December 4, 2023