boost.ai has returned the proxy control rights of AI to the customer experience team.
The new conversational AI platform version introduces large language model testing, AI assisted integration construction, and a fully configurable chat panel, all without the need to write any code.
Sannes, Norway, October 21, 2026 / PRNewswire / -- The enterprise-level conversational AI platform for regulated industries, boost.ai, today announced the release of version 14 of the platform, which brings a series of new features that enable customers ( CX ) and the AI operations team to directly control the building, testing of their AI agents, as well as the way they are presented to customers. This update will be gradually rolled out to customers over the coming weeks.
This release reflects a significant shift in enterprises' approach to managing conversational AI on a large scale. As AI agents take on more complex, transaction-based tasks, the teams responsible for their performance increasingly need to act swiftly—adjusting models, building integrations, and updating customer experiences at their own pace. This release is precisely crafted around this reality.
boost.ai CEO Jerry Haywood stated: "The issues that companies are facing now are no longer about whether conversational AI can handle complex transactional interactions; it can. The real challenge lies in whether the teams responsible for these interactions can move forward at the speed required by the business. What we do at boost.ai is to provide customers with tools to help them make better model decisions, so that AI investments match their needs, and to continuously control the evolution of their AI agents."
Test the trust model before relying on it.
The large language model (LLM) on this platform (Test Studio) now allows teams to run any use cases or safeguard mechanisms within its (AI) proxy on multiple (AI) models simultaneously, enabling them to accurately observe the performance of each model and make horizontal comparisons of quality, response speed, and cost.
The company stated that public benchmark tests measure the overall performance of models, but they do not reflect how well these models perform in specific use cases within particular organizations or under certain safeguard mechanisms. This update enables the CX team to identify, before facing customers, which scenarios can be handled just as well by simpler and less costly models, and which scenarios indeed require more powerful models.
As a result, the model selection approach will be more cautious: based on evidence rather than assumptions, and it will ensure that costs are aligned with the requirements of each task.
Build integrations without having to wait in line for developers.
boost.ai The platform has also made significant upgrades to the AI assistant that is directly built into its API Connector. The team can now describe the required integrations in natural language – from Salesforce and other CRM systems to automated platforms such as UiPath – and then provide the relevant API documentation to this assistant. Subsequently, they can gradually obtain functional integration code.
The company states that throughout the industry, integration has always been a historical bottleneck for transactional AI use cases. Querying balances, processing changes, or retrieving policy details all require connecting to backend systems, and traditionally, building such connections involved writing code – which also meant having to wait for the development cycle to complete. boost.ai API Connector Companion eliminates this dependency; teams can now build and optimize integrations on their own. The time required from proposing a requirement to completing the integration is equivalent to the time taken to describe those requirements, and no coding knowledge is necessary.
Configure customer experience as desired.
This update also includes a new version of the boost.ai chat panel, along with a redesigned configuration interface. The chat panel can now be used as a floating component, for a full-page experience, or directly embedded in a page, and all configurations are managed through the platform's user interface.
The team can maintain multiple configurations and apply each configuration to specific pages using URL. As a result, the same AI proxy can exhibit different behaviors at various stages of the customer journey, and these changes can be directly controlled by business users through the interface. New features include landing page views with suggested themes, as well as fixed areas for system messages and disclaimers. Together with the platform's existing capabilities, these features clearly indicate to end-users that certain responses are generated by AI to comply with the transparency requirements of the European Union's Artificial Intelligence Act.
boost.ai, Vice President of Product Development, Clay Cardozo stated: "In this release, we focused on the areas that customers have identified as being the most time-consuming for them. Model selection used to rely on guesswork, integration required coding, and interface changes needed front-end resources. Now, all three of these tasks have become something that the CX team can complete on their own, and they are all based on the same principle: the people who are closest to the customer experience should be the ones who shape that experience."
About boost.ai
boost.ai indicates that the company is a trusted leader in the field of customer experience solutions for regulated industries, driven by AI. Built around security, speed, and scalability, this platform enables rapid deployment, high resolution rates, and comprehensive mixed control through seamless integration of traditional natural language understanding (NLU) and large language models. The company claims to have completed over 650 successful deployments, boasts 600 online AI agents, and has achieved more than 150 million automated conversations in total. boost.ai states that the company helps global enterprises solve problems with greater confidence, achieve scaled automation, and trust every conversation. With its proven performance and enterprise-level reliability, boost.ai claims to have become a partner of leading global brands, including Nordea, Credit Union, of, Colorado, Sage, DNB, Trading, 212, and others. boost.ai also mentions that the company was rated as a leader in Gartner's "2025 Conversational AI Platform Magic Quadrant".
