Forethought
Forethought is an AI-powered customer support tool that helps teams automate responses and improve service efficiency, enabling faster resolution of customer inquiries.
Forethought is an advanced AI agent platform for customer support that automates responses, resolves inquiries end-to-end, and enhances agent efficiency, delivering faster resolutions and higher CSAT scores while reducing ticket volumes by up to 80%.[1][2][5] Key features include intelligent ticket triage based on customer intent, real-time agent assistance with contextual suggestions and autocomplete, omnichannel support across email, chat, voice, and more, plus AI-driven insights to detect knowledge gaps and optimize workflows.[1][2][4]
Its multi-agent system learns from company data, past tickets, and help center content to provide personalized, policy-compliant actions, seamlessly integrating with tools like Salesforce, CRMs, and knowledge bases without requiring replacements.[2][3][4] Designed for mid-size to enterprise teams in industries such as ecommerce, fintech, SaaS, healthcare, and education, Forethought excels in high-volume environments, scaling onboarding, handling complex regulations, and supporting use cases like order tracking, password resets, and compliance workflows.[1][3][5]
What sets Forethought apart is its agentic AI that reasons, decides, and acts autonomously—surpassing traditional chatbots by minimizing escalations, adapting dynamically to updates, and enabling up to 20% gains in agent productivity, as evidenced by customers like Upwork, Lime, Grammarly, and Cotopaxi.[2][4][5][6]
Forethought uses Aggregated Data (non-identifying, combined with other customers' data) for service improvement and algo
No training opt-out mechanism was identified in the legal documents.
Moderate risk. Suitable for internal data but avoid confidential information.
Forethought uses Aggregated Data (non-identifying, combined with other customers' data) for service improvement and algorithm development, but the language is somewhat vague about whether individual customer data feeds training. The DPA permits use of Aggregated/Deidentified Data for model improvement, but the distinction between what qualifies as 'aggregated' versus raw training data is unclear.
“[ToS] 'Forethought may use Customer Content and data...that (i) does not specifically identify Customer...and (ii) is combined with the data of other customers...for (a) maintaining, analyzing and improving the Services, including the algorithms underlying the Services'”
Forethought grants itself a broad license to use, copy, modify, and create derivative works of Customer Content 'solely to the extent necessary to provide the Services,' but also reserves the right to 'periodically access' content for 'quality assurance, benchmarking, technical support, or billing purposes.' The license scope is somewhat ambiguous and could permit uses beyond strict service delivery.
“[ToS] 'Customer hereby grants Forethought a non-exclusive, worldwide, royalty-free right to use, copy, store, transmit, modify, create derivative works of...Customer Content solely to the extent necessary to provide the Services...Forethought may periodically access Customer's account and Customer Content to improve, support, and operate the Services'”
Forethought shares data with 'contracted vendors, consultants, and other third parties' for IT, analytics, marketing, and customer support. A subprocessor list is published, and the DPA requires written agreements with subprocessors, but the language permits sharing with 'affiliates' and 'partners' with limited specificity about data protection controls.
“[ToS] 'Customer may request additional services from, certain third-party service providers in connection with Forethought's provision of the Services, and such third parties will have access to Customer's Confidential Information, including Customer Content and Customer Personal Data'; [DPA] 'Forethought may engage Subprocessors...Forethought will not engage new Subprocessors without giving Customer prior notice and a reasonable opportunity to object'”
The DPA states data will be deleted at the end of the Services or upon Customer request, but allows retention where required by law and for backup systems (which are 'securely isolated'). The Privacy Policy states retention is 'as long as necessary' but provides no specific timelines, creating ambiguity about actual retention periods.
“[DPA] 'At the end of the Services, or upon Customer's request, Forethought will securely destroy or return...the Personal Data...This requirement shall not apply to the extent that Forethought (i) is required by applicable law to retain some or all of the Personal Data, (ii) in relation to Personal Data archived on back-up systems'; [Privacy] 'Forethought will retain your personal information for as long as is necessary for the purposes it was collected'”
The ToS explicitly states that Customer Content remains the sole property of Customer and Forethought claims no right, title, or interest in Customer Content. The license granted to Forethought is limited to providing the Services and does not extend to ownership or sublicensing of outputs.
“[ToS] 'As between the Parties, Customer exclusively owns all right, title and interest in and to all Customer Content and Forethought claims no right, title or interest in the Customer Content'; 'Customer Content shall remain the sole property of Customer'”
The DPA references a security policy at forethought.ai/security-policy/ but the actual certifications are not detailed in the provided documents. The DPA incorporates Standard Contractual Clauses for GDPR, UK, and Swiss data transfers, and mentions CCPA compliance, but no SOC 2, ISO 27001, or HIPAA certifications are explicitly stated in the documents provided.
“[DPA] 'Forethought implements the following technical and organizational measures (including any relevant certifications) found at: forethought.ai/security-policy/'; [DPA] 'the SCCs shall apply...in relation to transfers of Personal Data subject to the GDPR'”
Track and measure your AI tools with Olakai's observability platform.
Explore Olakai