Squiz launches AI search tool for secure, compliant web answers
Squiz has announced the release of Squiz Conversational Search, a new feature aimed at providing AI-generated answers on enterprise websites using verified, organisation-owned content.
This solution is intended to modernise website search experiences by incorporating a conversational interface similar to those popularised by generative AI tools, while retaining strict content control for organisational governance and compliance. Squiz Conversational Search is integrated globally within the Squiz Digital Experience Platform (DXP) and utilises the Squiz Funnelback enterprise search engine.
Natural language interaction
Rather than displaying standard search result listings, Squiz Conversational Search allows users to pose questions in natural language and receive responses directly derived from content sets that have been pre-approved by the organisation. This approach aims to address the increasing user expectation for more intuitive and immediate access to information.
"This is about helping organizations meet users where they are. People don't want to dig through menus or search result pages anymore, they want to ask questions and get clear answers, fast," said John-Paul Syriatowicz, Founder of Squiz. "Squiz Conversational Search gives our customers a safe, smart way to embrace generative AI on their own terms, without giving up control of their content or user journeys."
The launch comes as a recent Google and IPSOS survey found that nearly 49% of Australians have used generative AI in the previous year, indicating a broad uptake of AI-driven information retrieval and heightened demand for interactive searching experiences. Squiz intends the new feature to provide enterprises with a secure way to leverage generative AI technologies on their own digital platforms, without relinquishing control over content accuracy or governance.
Technical underpinnings
The feature introduces a dual-agent model, where generative responses are produced and then independently cross-checked against the verified content repository before being delivered to the user. This system is built on Retrieval-Augmented Generation (RAG) architecture, which grounds AI-generated responses exclusively in sourced material that is managed by the organisation.
"Unlike generic AI chat tools, this is a product built for enterprise-grade accuracy and governance," said Greg Sherwood, Chief Technology Officer at Squiz. "Because it's powered by Squiz Funnelback, organizations define the exact collection of content responses are to be drawn from and can continuously monitor and improve accuracy."
Squiz states that this approach addresses the risk of AI hallucinations and ensures the reliability of answers presented to end-users. The underlying Squiz Funnelback engine is responsible for retrieving relevant documents from designated content sets and validation occurs before information is surfaced in the conversational interface.
Content governance and integration
Squiz Conversational Search is configured to allow client organisations to tailor both the source and nature of permitted content, supporting compliance and brand alignment requirements. According to Squiz, it integrates with existing technical environments without necessitating major architectural changes, and places an emphasis on accessibility and customisation.
The company has developed a consulting-led deployment approach designed to help organisations evaluate and prepare their content for use in AI-driven search, define their primary use cases, and implement conversational search capabilities in a manner intended to be controlled and strategic.
Squiz highlights the product's suitability for service-driven sectors where trust and transparency are critical, with the solution's validation and governance tooling targeted at supporting these priorities. The company has made the new feature available to all Squiz DXP customers and offers demonstration environments for those wishing to explore its capabilities before implementation.
