From predictive analytics to personalized fan experiences, data and AI are powering the next generation of motorsport, says Rohit Agnihotri, principal technologist at Infosys, and Dan Cherowbrier, CTIO of Formula E. When the ABB FIA Formula E World Championship launched its first race through Beijing's Olympic Park in 2014, the idea of all-electric motorsport still bordered on experimental. Batteries couldn't yet last a full race, and drivers had to switch cars mid-competition. Just over a decade later, Formula E has evolved into a global entertainment brand broadcast in 150 countries, driving both technological innovation and cultural change in sport. Gen4, that's to come next year, says Dan Cherowbrier, Formula E's chief technology and information officer. You will see a really quite impressive car that starts us to question whether EV is there. Formula E's digital transformation, powered by its partnership with Infosys, is redefining what it means to be a fan. "It's a movement to make motor sport accessible and exciting for the new generation," says principal technologist at Infosys, Rohit Agnihotri. From real-time leaderboards and predictive tools to personalized storylines that adapt to what individual fans care most about--whether it's a driver rivalry or battery performance--Formula E and Infosys are using AI-powered platforms to create fan experiences as dynamic as the races themselves. Technology is not just about meeting expectations; it's elevating the entire fan experience and making the sport more inclusive, says Agnihotri. AI is also transforming how the organization itself operates. Historically, we would be going around the company, banging on everyone's doors and dragging them towards technology, making them use systems, making them move things to the cloud, Cherowbrier notes.

The state-of-the-art code models, capable of understanding and generating code in numerous programming languages, are revolutionizing the way enterprises approach software development. With the ability to understand and generate code across a vast array of programming languages, these code models offer a significant boost in productivity. However, the one-size-fits-all approach of these generic multi-lingual code models often falls short in meeting the nuanced requirements of project-level coding tasks in an enterprise, which tend to be language-specific. This has led to the development of Narrow Transformers (NTs), specialized models further trained on a particular programming language, offering a more efficient solution for enterprises. These NTs are designed to optimize performance for a specific programming language, balancing the trade-offs between model size, inferencing cost, and operational throughput. As demand for tailored solutions grows, we can expect a surge in NT development, providing the precision and efficiency required by enterprise projects. However, in practice, the substantial economic cost associated with training and fine-tuning large code models renders language model experiments prohibitively expensive for most researchers and organizations.

In the context of enterprises accumulating proprietary unstructured data, AI-driven information retrieval solutions have emerged as vital tools for extracting relevant answers to employee queries. Traditional methods for developing such solutions often involve choosing between Retrieval Augmented Generation (RAG) or fine-tuned Large Language Models (LLMs). However, fine-tuned LLMs, comprising only generative models, lack a guarantee of factual accuracy, while RAG, comprising an embedding model and a generative model, assures factual precision (Lewis at al., 2020 [1]). Despite their superior performance in general, RAG based solutions often rely on pre-trained models, potentially leading to suboptimal alignment with enterprise-specific data. Addressing this challenge entails exploring two potential avenues: Firstly, recent studies such as RAFT (Zhang et al., 2024 [2]) explore the integration of fine-tuned generative models within a RAG pipeline to enhance accuracy, albeit requiring substantial domain-specific data to fine-tune the generative models. Alternatively, leveraging domain-specific embedding models within a RAG pipeline to enhance accuracy remains an underexplored area. Earlier efforts, such as BioBERT (Lee et al., 2019 [3]), SciBERT (Beltagy et al., 2019 [4]), and LEGAL-BERT (Chalkidis et al., 2020 [5]) have effectively demonstrated the efficacy of domain-specific embeddings in information retrieval tasks. These endeavors primarily investigated two methodologies: (a) extending the pre-training of BERT and (b) pre-training BERT from scratch, both employing domain-specific corpora. Despite yielding commendable results, these methodologies necessitated substantial domainspecific corpora, with figures as staggering as 21.3B words for BioBERT, 3.17B tokens for SciBERT, and 11.5GB of text data for LEGAL-BERT, thereby posing significant challenges, particularly in low-resource domains like enterprises.

Heading into 2023, a survey of 2,500 senior technology leaders and executives across thirteen industries finds that despite recent advances, most enterprise organizations are a long way from mastering artificial intelligence (AI). Conducted by the Infosys Knowledge Institute, the survey finds that 63% of AI models function only at basic sense (36%) or understand capability (27%), are not autonomous and often fall short on data verification, data practices and data strategies. Most organizations, however, are still new to AI with 81% of respondents deploying their first true AI system in only the past four years, with 50% deploying in the last two years. Only a quarter of respondents (26%) said they are highly satisfied with their data and AI tools, with the highest rates in the financial services sector outpacing all other industries by a wide margin, the survey finds. Nevertheless, the survey also suggests organizations can generate more than $460 billion in incremental profit if they improve data practices, trust in advanced AI and integrate AI with business operations.

Companies have been investing heavily in artificial intelligence (AI) systems, but all that work is not paying dividends. The digital giants, including the cloud giants and others such as Apple, Facebook, and Netflix, are able to convert data science to business value, but other large enterprises can't. They want to do AI, but they don't know how to get something out of it. What are the problem areas? Are companies new to AI, or do they use basic AI? Are there too high expectations from this niche technology?

Metaverse technologies are changing the game for manufacturers. From the way we make products to the way we manage them, there are some fundamental shifts underway thanks to AI, IoT and AR. The confluence of these technologies is enabling manufacturers to envision products and factories digitally before manifesting them physically. There are many interesting use cases in play already. The metaverse enables you to test as you build.

Scientists have theorized about the potential of quantum computing -- that is, a new approach to computation that uses probabilities, rather than binary signals, to make calculations -- for decades. But in recent years, both private and public sector investment into developing quantum computers has grown significantly, with one report projecting investments of more than $800 million in 2021 alone. Quantum technology could revolutionize everything from genomic sequencing to transport route optimization, from code-breaking to new materials development. But while quantum computers exist in the lab, general-purpose quantum computers aren't yet available for commercial use. How can businesses respond to potential disruptions from this technology before it has actually emerged into the mainstream market?

Management by humans can be dismal. "In the old world of cabbing, the drivers were often abused," says James Farrar, director of non-profit organisation Worker Info Exchange (WIE). Drivers would pay the same fee to drive for a taxi company, but receive differing amounts of business. "You'd have so-called'fed' drivers [fed with work] and'starved' drivers, with favoured drivers getting all the nice work," he says, with some dispatchers who allocated work demanding bribes. As a result, many welcomed dispatchers being replaced by algorithms: Farrar recalls cheering this in a session for new Uber drivers.

Countries such as China, the US, South Korea, and Russia are investing huge sums of money to develop their AI technologies. On the other hand, India is also picking up its pace and entering the race of AI. Artificial Intelligence is indeed the next big thing that would dominate the world. Even the top IT companies have now declared that AI is the future. Countries such as China, the US, South Korea, and Russia are investing huge sums of money to develop their AI technologies and other policy strategies related to the same. On the other hand, India is also picking up its pace and entering the race of AI.

"We call it single bottom-line sustainability, where I look at the single bottom line of all those elements, and I start attaching sustainability to it," Glickman says. "And I start looking at changes of value and then I can build a business case for change." As companies set sustainability goals--to be carbon neutral by 2050, for example--they're tackling complex challenges: regulations change, supply chains are complicated, especially during the current pandemic, and integrating new technologies into legacy systems is almost always a hurdle, technologically and culturally. Glickman suggests an incremental approach--he calls it micro change, embracing the fact that sustainability isn't a one-and-done paradigm shift. "These are things that can be done in a six-week period, eight-week period, that have tangible proof of concepts that can be measured, that can be done at different levels." Looking at current infrastructure investments, particularly in North America and Europe, as well as the increasing interest of stakeholders, the sustainability bar is expected to rise. "For the next three years you will see a lot of investment. You will see countries or businesses that want to be leading because they see an advantage," says Glickman. "Then you will see others have to move along in that direction also." This episode of Business Lab is produced in partnership with Infosys. Laurel: From MIT Technology Review, I'm Laurel Ruma, and this is Business Lab. The show that helps business leaders make sense of new technologies coming out of the lab and into the marketplace. Our topic today is sustainability, but on a global scale, from factories to supply chains to sustainable development goals for all the countries in the world. It's possible to design for sustainability, get a return on investment, and help fight climate change. My guest is Corey Glickman, who is the vice president and head of the sustainability and design business at Infosys. Corey is an expert in strategic design, digital transformation, customer experience strategy, and the use of visualization applied to the development of innovative products, processes, and services.