kama.ai Launches Trustworthy Hybrid AI Agents

Powered by GenAI’s Sober Second Mind®

TORONTO, May 07, 2025 (GLOBE NEWSWIRE) -- kama.ai, a leader in responsible conversational AI solutions, is pleased to announce the commercial release of the industry’s most trustworthy AI Agents powered by GenAI’s Sober Second Mind^®, the latest addition to its Designed Experiential Intelligence^® platform - Release 4.

The new Hybrid AI Agents combine kama.ai’s classic knowledge base AI, guided by human values, with a new enterprise Retrieval Augmented Generation (RAG) process. This in turn is powered by a Trusted Collection feature set that produces the most reliable and accurate generative responses. The Trusted Collection features provide pre-integrated intentional document and collection management with enterprise document repositories like SharePoint, M-Files and AWS S3 Buckets. Additional integrations will be made available upon request.

Optimizing the Human-AI Experience for Enterprise Efficiency

The diagram below shows how Hybrid Agents are informed by both kama’s 100% trustworthy Knowledge Graph AI, as well as generative AI enterprise RAG technology - based on the new Trusted Collection feature set. 

The process shows how Hybrid Agents improve over time with user feedback. Knowledge Managers and Subject Matter Experts (SMEs) review input to refine the system. This creates a positive cycle where responsible AI and human expertise work together to enhance conversations and automate enterprise processes.

Brian Ritchie, CEO and Founder of kama.ai says “We're responding to market demand to combine reliable enterprise approved content with the highest quality real-time generated responses. This approach grows enterprise knowledge faster - while maintaining maximum accuracy and brand control. Human-AI collaboration continues to prove essential for critical customer-facing enterprise solutions.”

AI chat interfaces using foundational models are popular for employee content creation that connect to large global and enterprise sources. This is a strong use case across the industry. However, kama.ai focuses on autonomous agents to both serve enterprises and their users with minimal risk. Release 4 strengthens this focus by adding a new generation of Responsible GenAI tools.

Josh Willard, Director of Marketing at Canadore College reports,

“The intelligence behind our ‘Panther’ AI Agent is kama.ai - and it shows. We’re seeing improved response times, better student engagement, and fewer emails to manage. It’s a win all around.”

Kaiyang Li, Head of Product at kama says, “Our platform now offers enterprises a multitude of options for internal and external AI Agents - from fully deterministic with zero risk, to hybrid applications with combined deterministic and generative capabilities. In hybrid applications, a generative response risk is mitigated with Trusted Collections, user notifications, and reference links.”

Designed Experiential Intelligence^® Release 4 helps enterprise experts work faster with greater ease. It generates draft responses automatically for a Knowledge Manager or SME to review. This is needed for highly sensitive applications (like HR), or for high volume customer facing applications. User inquiries, feedback, and AI drafts all help improve the system. Together, consumers, clients, partners, and SMEs create a more efficient and effective human-AI ecosystem.

GenAI’s Sober Second Mind^® API for Enterprise Builders

kama.ai Release 4 also introduces a new API supporting 3^rd party Hybrid AI Agent builders that can deliver 100% accurate and approved information curated for the enterprise. In these implementations, the kama knowledge graph AI delivers exact, deterministic enterprise information and process orchestration. This while the 3^rd party agent is free to deliver generated responses based on its own RAG and real-time governance capabilities. 

Enterprise RAG Knowledge Administration

Grounding RAG responses for greater accuracy is essential to delivering reliable answers to customers and employees. With Release 4, new enterprise RAG capabilities empower a new role of Knowledge Administrator within the platform. Knowledge Administrators can use pre-built connectors into enterprise document repositories such as SharePoint, M-Files, or S3 Buckets. These new connectors let Knowledge Administrators intentionally select the most appropriate enterprise documents for a knowledge collection. This group of sanctioned documents are then vectorized and published as Trusted Collections providing more contextually accurate generative responses. Beyond kama.ai’s Trusted Collections, Release 4 supports Responsible AI generation using models from OpenAI, Anthropic, and Google - with options to use custom enterprise models, as well.

By empowering enterprise Knowledge Administrators, Knowledge Managers and SMEs with Trusted Collections and trustworthy RAG capabilities, kama.ai continues its commitment to Responsible AI for enterprises. GenAI’s Sober Second Mind^® extends kama.ai’s traditional trusted knowledge graph AI technology. – As such, human oversight has now been extended into the world of Hybrid AI Agents for maximum flexibility in enterprise applications. It uses near real time human-AI collaboration which includes even more generative AI capabilities to provide a ‘best of both worlds’ scenario. kama.ai’s Release 4 gives the market the security of deterministic AI solutions, tied with the flexibility creative powers of generative AI solutions. All this done to give enterprises a brand-safe AI technology to both drive enterprise efficiency and user satisfaction.

About kama.ai

kama.ai is the provider of the responsible Designed Experiential Intelligence^® AI Agent platform. This platform provides trusted, and ethical responsible AI Agents to industries like finance, education, healthcare as well as cultural and community services. The platform prioritizes accuracy, is governed by human values, and eliminates or minimizes todays challenging issues of hallucinations or biases found in other AI systems. kama.ai delivers culturally sensitive and ethically sound AI solutions. ‘When it’s got to be right, it’s got to be kama.ai’

For media inquiries:

kama.ai: Charles Dimov, CMO

Phone: +1 (647) 702-1494

Email: [email protected]

A photo accompanying this announcement is available at https://www.globenewswire.com/NewsRoom/AttachmentNg/891a2495-5bf1-4638-842d-c404e6162028