By Mehab Qureshi
Artificial general intelligence (AGI) is often seen as the next evolution in AI technology, a future where machines don’t just perform tasks but understand and think like humans. AGI would not be bound to a single task or context— it would exhibit self-awareness, a sense of consciousness, and the ability to apply problem-solving skills across various domains. This contrasts with the AI of today, which excels in a specific set of tasks and relies heavily on massive data training and predefined algorithms to function.
Amidst this backdrop of technological pursuit, UAE-based QX Lab AI, founded by three Indians — Arjun Prasad, Tathagat Prakash and Tilakraj Parmar — has unveiled Ask QX — the first-of-its-kind hybrid AI system. This platform, rooted in both large language models (LLM) and neural network architecture, aims to seamlessly integrate AI into the everyday lives of Indians by supporting various languages, including native dialects. It plans to introduce two video and image products utilising advanced deep learning algorithms for
accurate data analysis and processing for enterprises, creative communities and individuals.
Decoding the tech
Akshay Golellu, an independent AI expert and analyst explains that LLM, such as GPT-3.5, are a type of AI developed through deep learning techniques, specifically within natural language processing (NLP). “LLMs are designed to understand, generate, and translate human language by being trained on vast textual datasets. These models analyse data to discern patterns, nuances, and language intricacies. Their size, determined by the number of parameters amounting to billions, equips them with the ability to process and produce human-like text. This allows them to perform tasks such as answering questions, creating content, summarising text, and even writing code,” he said.
Neural network architecture, on the other hand, is inspired by the human brain’s structure and function. It is composed of layers of interconnected nodes, or “neurons,” which process input data, such as images, sound, or text, and generate output through these connections.
“Each layer’s output becomes the subsequent layer’s input, with the final layer producing the overall output. Neural networks learn to perform tasks by considering examples, generally without task-specific programming. For example, in image recognition, they might learn to identify images that contain cats by analysing example images that have been manually labeled as “cat” or “no cat” and using this analysis to identify cats in new images,” Golellu added.
This architecture is fundamental in giving AI the ability to not just process, but also learn from data in various forms, contributing to deep learning – a key driver behind many of today’s AI advancements.
The breakthrough
The Ask QX platform is built upon a hybrid model—30% large language model and 70% neural network architecture — which is reflected in its capability to understand and interact with users with precision and contextual relevance. This is a value proposition that sets it apart from English-centric generative AI solutions dominating the market.
Tilakraj Parmar, co-founder & CEO of QX Lab AI, emphasises the platform’s focus on inclusivity, “Our vision for Ask QX is to bridge the existing gap, ensuring that the transformative potential of AI is accessible to all, not limited to a privileged few. The launch of Ask QX marks the initiation of a significant movement toward inclusive and accessible AI advancement in India.”
With a user base of over eight million users at launch, the platform supports more than 100 languages, featuring extensive Indian dialects such as Hindi, Bengali, Telugu, Marathi, Tamil, and several others. It also covers an array of global languages, like Arabic, French, and Japanese, making it a truly international offering.
Arjun Prasad, co-founder & chief strategy officer of QX Lab AI, noted, “We wanted to develop a product that would cater to users in every corner of India, in a language of their choice. The debut of Ask QX in 100+ languages, reaching eight million users already, stands as a testament to our unwavering dedication to creating a substantial impact.”
