Generative AI, or artificial intelligence, has the power to change how we live and work in so many ways; our creativity is the only limit.

At a recent roundtable discussion with Qlik entitled The Future of Data Analytics in the Age of Generative AI, I shared my thoughts about how the newly released foundation models in language or Large language models (LLMs) as we call them today are reshaping the work landscape. LLMs like GPT4, Claude etc., have been fine-tuned using reinforcement learning with human feedback to enable different categories of uses case, as listed below:

LLM as a language facade

Using LLMs as a layer of communication between humans and machines will make talking to software as easy as talking to another person. Instead of clicking through menus, you would just tell the software what you want and facilitate a seamless flow of information, transforming the way we interact with technology.

LLM as a co-pilot

Envision your digital sidekick enhancing your productivity exponentially, a testament to the possibilities of generative AI. LLMs could aid software programmers, supercharging their efficiency. Similarly, artists could use these models for inspiration, discovering new and creative ideas they haven’t thought of before.

LLM as a role-player

The intricate world model these LLMs acquire through training with trillions of words enables them to don any role and act in character. The power of role-play is limited only by our imagination — they could be coaches, companions, or even therapists.

LLM as an orchestrator

Moving beyond single-step interaction, LLMs could handle a series of tasks, making abstract interactions more concrete. Imagine your digital personal assistant breaking down complex processes into sub-tasks and diligently completing them step by step.

Also Read: How to stay creative in the age of Generative AI and Web3

While we are only at the inception of this transformation, several early use cases have already started to gain traction:

Q&A bot based on a data corpus

In the current search engine paradigm, information synthesis is a manual process. Generative AI models like ChatGPT bridges this gap, internalizing and summarising vast amounts of data, offering succinct and accurate responses. The technology eliminates redundant research, enabling users to devote their time to higher-value tasks. We can also engineer them to cite the sources of information they present to ground them.

Customer service bot

AI models can now imitate a range of communication styles and interact with customers in unique ways, allowing for the customisation of content and its delivery. They can change how they talk, their empathy, and their style based on how they want to talk to the customer.

This allows us to not only personalise what we tell the customer but also how we say it. The depth and reasoning behind every response can be engineered, taking customer service to an unprecedented level.

Coaching

AI coaching assistants can now provide an intelligent, interactive training experience. Be it for sales forces or for children learning new concepts, the models can role-play, ask follow-up questions, and provide feedback, offering a personalised learning experience.

In sales training, the AI bot can pretend to be a customer and ask good follow-up questions, pushing trainees to think about what the customer needs and how to sell to them. In education, these bots can act as personal tutors for kids, helping them understand what they’re learning.

Only the beginning

The power of generative AI lies in its ability to democratise data, bringing unstructured and structured data together to unlock business value in enterprises.

We are actively collaborating with companies like Qlik, for example, to see how we might augment and automate manual tasks involved with data management in order to help companies boost their productivity with quality and governance in mind.

Imagination and innovation would carry us to a future of work that we envisage. The key value of AI is in human augmentation – shifting employees and human labour to higher-value work.

However, a lot of engineering still needs to be done to put safeguards in place to make this technology more robust, safe, and fit for customer interaction.

You can watch my full presentation here.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. Share your opinion by submitting an article, video, podcast, or infographic

Join our e27 Telegram group, FB community, or like the e27 Facebook page

Image credit: Canva Pro

This article was first published on June 7, 2023

The post How to unlock new horizons with generative AI appeared first on e27.

Supply chain management software has evolved tremendously over the last decade. With cutting-edge technologies like artificial intelligence (AI) and machine learning being incorporated, these solutions are getting smarter and more intuitive every day.

In this article, we’ll look at how AI and ML are changing the game for supply chain management software and enabling organizations to make better and faster data-driven decisions.

Smooth supply chain management software is key for business success in today’s complex and constantly shifting markets. Companies need to manage their supply chain well to deliver products and services while keeping optimal inventory levels.

This means collecting and analysing tons of data on suppliers, production, inventory, transportation, sales, and more. Trying to make sense of all that information manually is challenging and time-consuming.

This is where AI and ML come in super handy. They give companies advanced tools like demand forecasting, inventory optimisation, supplier relationship management, and logistics routing to uncover patterns and insights from complex data to optimise planning and operations in the supply chain. With supply chain management support from AI and ML, businesses can streamline operations, reduce costs, and better serve customers.

AI and ML for more accurate demand forecasting

Accurate demand forecasting is crucial for efficient inventory and production planning. Old-school forecasting relied on statistical methods like moving averages. However, these have limitations in identifying complex nonlinear patterns. AI and ML models can detect intricate relationships and patterns in historical sales data much better. By analysing bigger datasets with more parameters like promotions, pricing, seasons, events, etc., they provide super accurate demand predictions.

ML techniques like neural networks can continuously learn from new data. This allows real-time refinement of forecasts in response to emerging trends. Companies can react faster to changes in customer preferences. AI also enables automated monitoring of forecast accuracy and exception handling for products with unusual trends. Instead of relying on fixed formulas, AI-enabled systems continuously optimise algorithms and models based on results.

Also Read: Hacking customer engagement in Indonesia’s agri supply chain

For example, if demand rises during holiday seasons, ML models can factor this in automatically. As new products are launched or old ones are discontinued, the system adjusts estimations seamlessly. This level of automation and flexibility is impossible to achieve manually.

Smarter inventory optimisation

Keeping optimal inventory is crucial for customer service and working capital management. Too much stock leads to higher carrying costs and obsolescence risks. Too little causes lost sales and backorders. ML algorithms can consider fluctuating demand, supply uncertainties, logistics delays and other constraints to determine ideal stock levels across the network.

AI can also improve inventory productivity by automating warehouses. Computer vision guides autonomous robots to locate and move inventory efficiently. This accelerates order processing, improves accuracy and allows 24/7 operation.

For example, smart inventory optimisation systems can monitor shelf life, seasonal demand shifts, waste reduction goals and other factors to align stock with business objectives beyond just costs. AI enables leaner, flexible and eco-friendly inventory management.

Dynamic supply chain network optimisation

SCM software with AI capabilities can dynamically optimise supply chain networks in response to changing conditions. The AI engine processes massive data on costs, lead times, risks, transportation lanes, sourcing options, duties, exchange rates, etc. It then uses advanced algorithms to determine the optimal locations and capacity of suppliers, factories, warehouses, cross-docks and outlets to minimise costs and maximise service levels.

As conditions change, the system reruns simulations and adapts the supply chain network for optimal performance. Manual design of such complex global networks can take months. But AI-powered software can crunch through numerous scenarios in minutes to optimise the supply chain in real-time.

For instance, weather delays, port congestions or other disruptions can frequently alter transportation costs and lead times. AI enables shifting supply paths dynamically to maintain continuity at the lowest cost. Sudden demand surges can be met efficiently by recalibrating inventory deployment and capacities with AI’s help.

Smarter sourcing and procurement

AI is transforming sourcing and procurement through automation and data insights. For example, routine tasks like issuing RFQs, analysing bid responses and preparing contracts can be automated using AI. Chatbots allow natural language interactions to quickly address supplier queries.

Big data analytics uncovers trends like price changes, supply risks, quality issues, etc., to support strategic sourcing decisions. AI determines the right procurement strategies for different spending categories based on value drivers instead of reactive buying. This brings major savings with better supplier terms and reduced maverick spend.

Also Read: Enhancing cyber supply chain resilience: A vision for Singapore

AI can also analyse negotiations with suppliers to continuously improve negotiation strategies and outcomes. It provides insights into which suppliers have higher negotiation room or where bundling spending could get better terms. This allows systematic optimisation of value.

Proactive supply risk management

Supply disruptions can wreak havoc on businesses. ML applies sophisticated pattern recognition and probabilistic modelling on news feeds, financial reports, weather data, transport records, etc., to identify likely disruption causes like natural disasters, trade wars, production issues, strikes, etc.

It analyses the potential impact on capacities, lead times and costs across the network. AI simulation helps mitigate risks proactively through safety stock optimisation, alternate suppliers, route changes, etc. Instead of reacting to disruptions, organisations can get ahead of problems.

AI-driven supply risk management also enhances transparency across tier-two, three and lower-tier suppliers to uncover hidden risks. It enables building contingency plans and scenarios to handle disruptions smoothly. This minimises downtime and customer impact.

Continuous process improvements

AI tracks all supply chain processes and exceptions. It analyses the root causes of inefficiencies like long lead times, quality problems, inaccurate planning, stockouts, etc. AI also estimates the cost impact of process bottlenecks. It uses computer vision to monitor process adherence on factory floors.

The insights allow focused process improvements to enhance productivity. AI also alerts when critical process parameters exceed limits. This enables proactive troubleshooting before issues arise. The continuous feedback cycle sustains gains over the long run.

Also Read: #dltledgers unveils 2023 trends in supply chain digitisation

For instance, AI can track invoice processing times, identify delays from missing information or workload spikes and reroute tasks automatically to improve turnaround. It can adjust warehouse staffing based on order volumes to maintain speed. AI enables self-optimising supply chain processes.

The future with AI and blockchain

Blockchain provides secure, transparent distributed ledger technology to improve end-to-end supply chain traceability. Combining it with AI and ML unlocks more value. AI can analyse blockchain transactions to uncover patterns, risks and insights. Smart contracts enabled by blockchain allow automated workflow execution.

Blockchain establishes a unified data source across networks. AI analyses this data for continuous optimisation. Smart contracts automate execution without conflicts. Together, they enable seamless cross-organisation integration while ensuring trust, security and compliance. This next-gen supply chain architecture minimises inefficiencies, disputes and disruptions.

Final thoughts

With capabilities like machine learning, computer vision and natural language processing, AI-powered SCM solutions help companies achieve new benchmarks in speed, accuracy and efficiency. By leveraging the convergence of AI and blockchain, future supply chains will become intelligent, self-learning networks that maximise value. The possibilities are exciting as AI and ML progress rapidly. Companies that embrace these technologies today will gain a real competitive edge.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. Share your opinion by submitting an article, video, podcast, or infographic

Join our e27 Telegram group, FB community, or like the e27 Facebook page

Image credit: Canva

The post Leveraging AI and ML in supply chain management for smarter decision making appeared first on e27.

In today’s ever-evolving entrepreneurial landscape, the notion of startup success has transcended the boundaries of Silicon Valley. Aspiring entrepreneurs are no longer limited by geography, and their dreams of building globally successful ventures are now challenging Silicon Valley’s historical dominance as the go-to destination for startups.

Techstars itself began with three simple ideas: entrepreneurs create a better future for everyone, collaboration drives innovation, and great ideas can come from anywhere. So supporting founders looking to create successful startups globally is a mission we have long been engaged in.

Embracing new horizons: The potential of anywhere startups

It has never been easier to launch a startup. Advancements in artificial intelligence and technological tools have significantly simplified the process of starting a business, making it more accessible for entrepreneurs. For instance, the rise of e-commerce platforms has minimised the need for physical stores, enabling businesses to operate online and reach a global customer base.

Digital marketing tools have made it easier to promote products and services, targeting specific audiences with precision. Cloud computing has revolutionised data storage and collaboration, reducing infrastructure costs and facilitating remote work.

Even the entrepreneur’s ability to raise business capital has been somewhat democratised by crowdfunding platforms that give them access to a range of sponsors and investors. These technological advancements have significantly reduced barriers and empowered entrepreneurs to pursue their business ideas with greater ease and efficiency from anywhere in the world.

The untapped potential of ‘anywhere Startups’ has been further reinforced by the COVID-19 pandemic. The advent of remote work and enhanced access to resources have given rise to vibrant startup ecosystems in unexpected corners of the world, making entrepreneurship a global phenomenon.

According to a survey report by McKinsey & Company, the global pandemic accelerated the digitisation of customer interactions with companies by several years.  At the height of the pandemic, several startups were still successfully launched and operated outside of Silicon Valley, showing entrepreneurs that success knows no geographic limits.

With the internet as their powerful ally, entrepreneurs now have the ability to connect with customers, investors, talents, and mentors on a global scale. This newfound freedom empowers them to pursue their dreams and build successful ventures beyond the traditional confines of Silicon Valley.

The role of geography and community

Geography still plays a pivotal role in shaping startup success. Each region possesses unique strengths, resources, regulatory climates, challenges, and market demands that must be navigated by entrepreneurs.

Also Read: Echelon: How increased emphasis on ESG elements in fund management will affect early stage startups

To maximise these strengths and navigate the challenges effectively,  it is necessary to fan the flames of entrepreneurial collaboration and startup community engagement within the region. This is the very reason several governments across the globe are now paying attention to promoting initiatives that bring entrepreneurs together and grow their local ecosystems.

A good example of such initiatives is the Anjal Z Techstars founder catalyst program which is a partnership between the Abu Dhabi Early Childhood Authority and Techstars to help edutech startups from across the globe get localised to Abu Dhabi.

To further support the regional development of startup communities globally, Techstars also offers a startup community catalyst that is a combination of multiple programs aimed at igniting and scaling startup communities from the ground up in partner regions. The fostering of local entrepreneurial communities sparks innovation and collaboration.

It can also create a nurturing environment that provides easier access to talent, support, and industry-specific knowledge relative to the region.  By understanding the local landscape and building collaborative communities, entrepreneurs in these regions can better leverage their geographic advantages and key into untapped opportunities.

Pre-accelerators and mentorship: Nurturing the entrepreneurial potential

From the initial spark of an idea to the eventual launch of a product and beyond, the entrepreneurial path is filled with challenges and uncertainties. Many founders usually venture in not knowing what to expect. However, pre-accelerator programs and mentorship can help bridge the gap and provide support during these crucial early stages. 

As the world’s largest pre-seed investor,  Techstars knows firsthand the value of mentorship to early-stage entrepreneurs, and that is why our community programs, such as Startup Weekends and Founder Catalysts, are meticulously designed to help founders with the necessary mentorship and support they need through every milestone of their early entrepreneurial journey- that is, from refining their ideas and defining their value proposition to preparing for future investments in our accelerator programs. 

Techstars collaboration: Fostering startup diversity

Diversity and inclusion have become imperative in the startup industry, and Techstars actively collaborates with entrepreneurs from diverse backgrounds and regions to foster a more inclusive ecosystem.

Entrepreneurs and partners that collaborate with Techstars are exposed to global networks and funding opportunities, as well as a supportive community of diverse mentors, entrepreneurs, and investors. These interactions can break down cultural and regional barriers. It also causes founders and partners to be open-minded and embrace global perspectives.

Also Read: Empowering startup entrepreneurs: Harnessing benefits of Web3

A community of diverse entrepreneurs brings fresh ideas, cultural insights, and innovative solutions to the table. The collective expertise and experience shared by mentors and peers empower entrepreneurs to challenge the status quo, disrupt industries, and build scalable startups.

Building the entrepreneurs and thriving communities of the future

Throughout our quest to support founders, we have found that a collaborative and strategic approach is always required when building startup communities. To build the entrepreneurs of tomorrow, we must first start by empowering the children and youth of today, not just in the US.

We can do this by prioritising entrepreneurial education, whether in the form of pre-accelerators, accelerators, and Tech hubs or actively in our schools and universities. This will help young individuals develop an entrepreneurial mindset and equip them with the skills and knowledge needed to navigate the startup landscape. Strengthening our collaborations between academia, industry, and government can drive research and development, further encouraging innovation and breakthrough technologies.

Additionally, fostering a culture of risk-taking and embracing failure as a learning experience is also essential for entrepreneurial growth. Promoting diversity and inclusivity within the entrepreneurial ecosystem is also key to unlocking new perspectives and driving innovation.

Lastly, continuous support and investment in emerging technologies and industries will help create thriving entrepreneurial hubs that shape the future of economies and industries.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. Share your opinion by submitting an article, video, podcast, or infographic

Join our e27 Telegram group, FB community, or like the e27 Facebook page

Image credit: Canva Pro

This article was first published on June 8, 2023

The post Embracing global entrepreneurship: Redefining startup success beyond Silicon Valley appeared first on e27.

In the dynamic world of Web3 startups, understanding and implementing effective financial models is crucial for achieving long-term success. The emergence of blockchain technology and decentralised finance (DeFi) has revolutionised the way startups operate, presenting unique challenges and opportunities. To navigate this rapidly evolving landscape, entrepreneurs need to adopt innovative approaches to financial planning and management.

In this comprehensive guide, we will explore the guiding principles for developing robust financial models tailored specifically to Web3 startups. From revenue streams to token economics and risk management, we will delve into the key aspects that drive financial success in this exciting domain.

Understanding the Web3 landscape

Before diving into the intricacies of financial modelling for Web3 startups, it is essential to have a comprehensive understanding of the fundamental concepts that define the Web3 landscape.

By familiarising themselves with decentralised finance (DeFi), non-fungible tokens (NFTs), smart contracts, and other essential components of the Web3 ecosystem, startups can align their financial models with the specific dynamics of the decentralised world.

Decentralised finance (DeFi)

Decentralised finance, or DeFi, refers to the use of blockchain technology and smart contracts to create financial applications that operate without intermediaries. Traditional financial services such as lending, borrowing, trading, and asset management are redesigned and decentralised, offering increased transparency, security, and accessibility to users. In the Web3 ecosystem, DeFi protocols enable startups to develop innovative financial products and services while removing traditional gatekeepers.

Web3 startups should explore various DeFi applications, including decentralised exchanges (DEXs), lending platforms, yield farming, and liquidity provision. By understanding the mechanics and potential risks associated with these platforms, startups can strategically incorporate DeFi elements into their financial models, leveraging the benefits they offer while mitigating any associated risks.

Non-fungible tokens (NFTs)

Non-fungible tokens, or NFTs, have gained significant attention in the Web3 world. NFTs are unique digital assets that can represent ownership or proof of authenticity for a wide range of digital and physical items, such as artwork, collectibles, virtual real estate, and more. NFTs are typically built on blockchain platforms like Ethereum, allowing for verifiable ownership and provable scarcity.

Also Read: Sony & UMG join forces with Snowcrash to revive NFTs: Here’s why the digital trend is far from dead

For Web3 startups, NFTs present an exciting avenue for monetisation and user engagement. By incorporating NFTs into their financial models, startups can explore revenue streams such as NFT sales, licensing, fractional ownership, and royalties. Understanding the dynamics of NFT markets, including trends, valuations, and user preferences, will be crucial in designing effective monetisation strategies.

Smart contracts

Smart contracts are self-executing contracts with the terms of the agreement directly written into code. These contracts automatically execute predefined actions when specific conditions are met, eliminating the need for intermediaries and enhancing the security and efficiency of transactions. Smart contracts are a fundamental building block of the Web3 ecosystem, enabling a wide range of applications, including decentralised exchanges, decentralised finance protocols, and more.

Web3 startups should grasp the concept of smart contracts and their potential applications. By leveraging smart contracts in their financial models, startups can automate processes, reduce costs, and ensure trust and transparency in their operations.

Understanding the programming languages used for smart contract development, such as Solidity, and the associated best practices will be essential for startups seeking to harness the full potential of this technology.

Web3 ecosystem and interactions

In addition to DeFi, NFTs, and smart contracts, there are numerous other components within the Web3 ecosystem that startups should be familiar with. These include decentralised storage solutions, identity management systems, oracle services, governance mechanisms, and more.

Understanding the interactions and dependencies between these components will enable startups to design financial models that account for the broader Web3 infrastructure and the potential synergies it offers.

By comprehending the dynamics of the Web3 landscape, startups can leverage the power of decentralised technologies in their financial models. This understanding will allow them to identify relevant revenue streams, incorporate token economics, assess risks and opportunities, and make informed decisions that align with the unique challenges and opportunities of the decentralised world.

Principles of financial modelling for Web3 startups

Understanding blockchain economics

Web3 startups are built upon blockchain technology. The financial model for such startups must reflect an understanding of the underlying blockchain economics. Factors like gas fees (transaction costs on a blockchain), mining rewards, and tokenomics (economic system around the token of a specific blockchain) will have significant implications on the startup’s financial dynamics.

Incorporating tokenisation

Web3 startups often use tokens as a mode of value exchange within their ecosystem. These tokens can serve various functions like utility tokens (providing users with access to a product or service) or security tokens (representing ownership in an asset). Their volatility in value needs to be factored into financial projections, and possible capital gain scenarios must be accounted for.

Handling regulatory uncertainty

Given the relatively novel nature of Web3 and the ensuing regulatory ambiguities, startups in this domain need to model the potential financial impacts of regulatory changes. This could include costs for compliance, penalties, or changes in user behaviour resulting from such regulatory decisions.

Forecasting user growth

User adoption and growth are vital to Web3 startups, with direct implications on financial performance. The financial model should consider different growth scenarios and examine the corresponding impacts on revenues and costs.

Accounting for network effects

The value of Web3 startups often grows as the network expands. This phenomenon, called network effects, should be incorporated into financial projections, including the impact of growth on value and costs.

Building a financial model for a Web3 startup

Now, let’s walk through a simplified version of building a financial model for a Web3 startup.

Revenue estimation

For most Web3 startups, revenues may come from transaction fees, staking rewards, or selling tokens. It’s crucial to forecast revenues based on estimated growth, token value changes, and market dynamics.

Also Read: Web3 startups: The next big thing investors are flocking to

Cost projection

On the expense side, typical costs include development, operations, and marketing. Additionally, costs unique to Web3, like gas fees or smart contract audits, must be accounted for.

Financial statements

Build the traditional profit and loss statement, balance sheet, and cash flow statement. However, these will likely need modifications. For example, balance sheets might need to include token reserves, while the cash flow statement needs to account for cryptocurrency flows.

Scenario analysis

Given the volatility and uncertainty in the Web3 space, it’s critical to model different scenarios to understand potential outcomes and risks.

Valuation

Valuing a Web3 startup is challenging, given the scarcity of comparable companies, token price volatility, and regulatory risks. Techniques like Discounted Cash Flow (DCF), token economy valuations, or using multiples from a few existing similar companies can provide some guidance.

Final thoughts

Developing robust financial models is essential for the success of Web3 startups. By understanding the principles of financial modelling specific to the Web3 ecosystem, entrepreneurs can make informed decisions, attract investors, and navigate the challenges and opportunities in this dynamic landscape.

Incorporating elements such as blockchain economics, tokenisation, regulatory considerations, user growth forecasting, and network effects will enable startups to build comprehensive financial models that drive sustainable growth and long-term success. Through diligent research, analysis, and scenario planning, Web3 startups can optimise their financial strategies and position themselves for success in this exciting and rapidly evolving domain.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. Share your opinion by submitting an article, video, podcast, or infographic

Join our e27 Telegram group, FB community, or like the e27 Facebook page

Image credit: Canva Pro

The article was first published on June 6, 2023

The post Financial models for Web3 startups: Guiding principles for success appeared first on e27.