AI-powered SaaS platforms and decentralised networks are transforming sustainability in the fashion industry by enhancing supply chain transparency and automating carbon emissions tracking.
The intersection of deep tech and compliance infrastructure, enabled through AI-powered SaaS platforms and decentralised AI platforms, is restyling multiple facets of the fashion industry. The impact ranges from design and production to sales and marketing, with a particular focus on e-commerce and retail channels.
In the fashion industry, overseeing a complex supply chain with multiple stakeholders often results in inefficiencies and a lack of transparency. Many fashion brands struggle to track the environmental impact or labour conditions throughout their entire supply chain, especially when it comes to decentralised, lower-tier suppliers.
AI SaaS platforms enable fashion companies to automate the tracking and reporting of carbon emissions, with a focus on Scope 3 or indirect greenhouse gas emissions that occur within the supply chain. By collecting data from suppliers and manufacturers, these platforms provide real-time insights into emissions levels, streamlining the reporting process and enabling companies to meet regulatory requirements and industry standards with minimal manual effort.
However, Narendra Makwana, co-founder and CEO of GreenStitch, a sustainability software tailored for the fashion and textiles industry, said that assessing a company's emissions requires analysing extensive data. Large textile and fashion companies manage millions of purchase orders each month, making data entry prone to errors. For instance, the term 'cotton' might be mistakenly entered as 'COD' or 'CEO', leading to unstructured data. This is where machine learning can help by organising this data effectively.
Makwana, in conversation with AIM, highlighted the benefits of applying machine learning in emission analysis. GreenStitch's AI-powered SaaS platform is trained over time to handle unstructured data from multiple companies, making the process faster and more efficient.
The company emphasises third-party verification of its modelling process. Customers provide data, such as bills of materials, which is then verified using Optical Character Recognition (OCR) technology. If a customer's input, such as 1,000 kilowatt-hours of electricity, does not match their bill of 5,000 kilowatt-hours, the system flags the discrepancy and notifies them.
"While we offer some assurance on our modelling, it's the customers' responsibility to ensure their inputs are accurate. We also use advanced AI analytics to identify discrepancies. Our final assurance focuses on confirming the correct alignment of emission factors, methodologies, and frameworks, validated through third-party verification of our code," the CEO said.
India stands out as a leading exporter of fashion and textile goods, particularly benefiting from the 'China Plus One' strategy, wherein companies diversify their manufacturing and sourcing operations by expanding beyond China to reduce dependence and mitigate risks. Consequently, the local demand for manufacturing and producing fashion products has increased, prompting a greater focus on integrating SaaS platforms into day-to-day operations.
GreenStitch's automation efforts focus on integrating data from operational systems, such as Supervisory Control and Data Acquisition (SCADA) and Enterprise Resource Planning (ERP), which enables the elimination of manual data entry. By efficiently structuring this data, they provide valuable insights, such as climate scores. In cases where companies lack effective data workflows, their platform helps capture the necessary information efficiently.
GreenStitch offers templated workflows and facilitates the generation of various reports -- product, buyer, and active level -- through AI, resulting in visually appealing outputs complete with graphs and commentary. Companies like Aditya Birla Fashion and Retail often need to comply with multiple frameworks such as the Business Responsibility and Sustainability Reporting (BRSR), Global Reporting Initiative (GRI), Customer Data Integration (CDP) and Geostatistical Software (GS+).
Shashank Sripada, co-founder and COO of Gaia, told AIM that creating a network of agents that monitor various aspects of a car, such as emissions, usage, and wear, can unlock significant opportunities, much like those in the clothing industry. This setup enables the automatic collection and analysis of data from vehicles and other emission sources, which can be fed into emissions maps. Such insights can enable policymakers to make better decisions.
Moreover, by incorporating economic, social, and other relevant variables alongside emissions data, we can develop tailored decarbonization recommendations through AI, Sripada added.
In India, we often rely on outdated databases from Western countries.
Makwana suggests that the government could publish its own data based on recent research, which would benefit local companies and enhance transparency. For example, the EcoInvent database, last published in 2015, presents an inaccurate view of India compared to Bangladesh, despite India's significant advancements in renewable energy since then. Updating these databases would better reflect our progress and showcase our lower emissions, he added.
Decentralised AI networks could empower individual nodes, such as suppliers, manufacturers, and even consumers, to independently collect, process, and share data about the sustainability practices or labour conditions at each point in the supply chain. Instead of relying on a centralised authority to monitor this, each participant could retain ownership of their data while contributing to a shared decentralised platform for sustainability and compliance.
Sripada said that he sees a shift in ownership impacting industries and businesses in much the same way that early websites on the internet evolved, moving from being hosted on platforms like GeoCities to owning and controlling their websites with complete flexibility.
According to him, centralised models are viewed as a gateway to the future of agents, but they are just one part of the equation. Both existing and new businesses are likely to adopt open-source, decentralised models and agents, benefiting all involved. This transition will also be advantageous for centralised LLM providers, as more individuals will engage with niche agents and models they can create.
AI enhances customer experiences and business operations, particularly in retail and e-commerce. By analysing extensive consumer data with machine learning algorithms, retailers create advanced personalisation, recommendation, and automation features. These improvements boost efficiency and customer interactions, driving increased revenue for companies.
Sripada believes that, like supply chain management, personalised data on fashion preferences will enhance real-time clothing and fashion advice. "Just as we saw disruption in fashion retail with the rise of D2C, Shopify-led online outlets, I believe AI will empower fashion creators to design, market, distribute, and track their D2C businesses at scale even more effectively, further narrowing the gap between them and fashion houses like LVMH, whose marketing and distribution networks are arguably more dominant than their creative edge."
He highlighted that AI offers individual designers the tools of a large fashion house, enabling efficient operations and personalised customer interactions through virtual twins. "It truly is an exciting time," Sripada concluded.
