PointAI Unveils Physics-Based Simulation AI for Faster, More Accurate & Affordable Digital Fashion Commerce

Uttar Pradesh, India, August 20th, 2026, FinanceWire

PointAI, an AI company specialising in patented Simulation AI for commerce and enterprise, has launched a real-time virtual try-on solution for apparel and fashion accessories. The company says its technology delivers significantly greater accuracy, affordability and speed compared with commercially available GenAI-based solutions. PointAI has also unveiled its AI platforms for e-commerce & enterprises and plans to launch an agentic commerce app for fashion shopping across India, Japan, the US, the UK and Singapore.

Unlike conventional GenAI systems that primarily predict pixels, PointAI's Simulation AI combines generative models with physics-based simulation to replicate how fabrics behave on different body shapes and sizes. The technology is designed to create highly personalised, near-real digital representations of apparel in fractions of a second.

Challenges in GenAI Based Virtual Try On

Enterprise AI continues to face significant challenges in moving from pilots to production. RAND Corporation estimates that more than 80% of AI projects fail, while IDC research has found that 88% of AI proofs-of-concept do not reach production. MIT's Project NANDA has also reported that 95% of generative AI pilots deliver no measurable return.

In virtual try-on, a key challenge is that GenAI systems can struggle to accurately account for individual body characteristics such as bust, waist, hips, torso proportions, and leg length. Fabric texture, colour, and design details can also vary between generated outputs. Google's own developers' forum documents ongoing problems with stability, face distortion, and incorrect rendering. Commercial try-on APIs can cost approximately $0.15 per image and take 30-60 seconds to render.

Physics Not Pixels

PointAI's patented Simulation AI models simulate garments as physical objects governed by the underlying physics of fabrics. This is combined with deep-learning models trained across more than 200,000 body-type variations, enabling the system to render realistic fabric on a human body in under one second, at approximately 1/100th the cost of GenAI alternatives.

Yann LeCun, the Turing Award-winning scientist and former Chief AI Scientist at Meta, has argued that today's leading AI models remain fundamentally limited in their ability to reason about the physical world. He has also highlighted the limitations of approaches that rely primarily on predicting pixels, rather than building an understanding of how the physical world works.

Nitin Vats, Founder & CEO of PointAI, said, “Asking a GenAI model to generate an image by predicting every single pixel of how fabric drapes on a human body will not be as accurate as the image generated by simulating the underlying physics of the fabric. Also, it will use more computational resources and time to produce results in comparison to Simulation AI.”

Enabling Agentic Commerce

McKinsey estimates that agentic commerce could represent a $3-5 trillion opportunity by 2030. For AI agents to browse, evaluate and transact on behalf of consumers, accurate and instant product visualisation will become increasingly important.

PointAI is developing a dedicated agentic commerce platform for fashion and accessories, combining a real virtual try-on, mix-and-match capabilities, an AI fashion advisor and other shopping tools. The company plans to eventually extend the platform to categories including automobiles, home appliances and gadgets.

Its enterprise platform will offer e-commerce brands virtual try-on, fitting and recommendation plugins, image-based search, a virtual stylist, catalogue generation and customer support solutions. PointAI has also patented solutions for image, video and 3D generation. The company is also training a ~250B parameter GenAI model for solving all kinds of conversational needs of enterprises.

Nitin Vats, Founder & CEO of PointAI, said, “The last decade of commerce was built for people who browsed. The next is built for people who delegate to the agents and assistants who can provide a real showroom experience online.”

Proven in the Market

Amazon, Rakuten, Flipkart, Myntra, ABFRL and others have already been clients of PointAI. The company serves brands across India, Japan, South Korea and the US. At Bharat Tex 2026 in New Delhi, PointAI launched an in-store virtual trial standee with Gokaldas Exports, showcasing apparel styles produced for global brands including Adidas, Aerie, Banana Republic, Free Assembly, M&S and Puma.

Relevance to European Retail

Physics-based AI could be particularly relevant to European retail, where explainability and compliance are becoming increasingly important under the EU AI Act. A deterministic system based on physical rules can offer greater traceability than opaque generative outputs.

Accurate, body-specific virtual fitting could also help address fashion returns and textile waste by enabling consumers to make more informed purchasing decisions before buying.

About PointAI

PointAI Technologies is an AI company, specialising in patented Simulation AI solutions for retail and digital commerce.

Founded by Nitin Vats, the company has built a portfolio of more than 70 granted global patents spanning virtual try-on, AI catalogue generation, digital twins, interactive 3D commerce, AI-powered shopping assistants, and retail automation. Its foundational patents have received more than 150 patent citations, including from AT&T Bell Labs, reflecting the strength of its intellectual property.

Nitin Vats is a former researcher at Microsoft Research Lab and is recognised for his contributions to applied mathematics and artificial intelligence, including work on high-speed computation, approximation algorithms, and World records in Mental mathematics.

PointAI is backed by leading global investors, including Lip-Bu Tan, CEO of Intel. Its patented Simulation AI powers commerce experiences across platforms.

Contact