PNN
Bengaluru (Karnataka) [India], February 3: India's AI ecosystem may have just taken a global leap. Bangalore-based startup Assessli has launched the Large Behavioural Model (LBM), a domain-agnostic foundational infrastructure designed to understand human behaviourgoing beyond conventional AI models that focus only on language or past actions.
Founded by Suraj Biswas, Aruna Dey, and Saurabh Gupta, Assessli aims to fill a key gap in the AI landscape. While existing models like GPT and Gemini interpret language or search patterns, they do not capture why humans behave the way they do. LBM, by contrast, processes over 1 trillion behavioural data points and achieves 99.6% personalisation accuracy, enabling AI systems to respond to emotional cues, decision patterns, and daily rhythms in real time.
The model supports diverse applications across industries, including personalised medicine, digital twins, AI companions, cognitive work agents, neurofeedback environments, subconscious commerce, and robotics. The company has now announced Dots-In its consumer-facing brand and the first proof-of-concept running natively on LBM set to launch in the coming months.
LBM represents a horizontal AI layer, offering a unified infrastructure for human understanding that can integrate with multiple platforms and devices globally. Assessli is also in advanced talks with a leading smartphone manufacturer to embed LBM across its ecosystem, potentially bringing behavioural intelligence to hundreds of millions of users.
As India stakes its claim in the global AI race, Assessli's LBM stands out as a foundational infrastructure developed in India, designed to complement existing AI systems and provide unprecedented insight into human behaviour.
Early access and more information: dotsin.ai
(ADVERTORIAL DISCLAIMER: The above press release has been provided by PNN.will not be responsible in any way for the content of the same.)
Disclaimer: This post has been auto-published from an agency feed without any modifications to the text and has not been reviewed by an editor