GenAI and the Way Forward

When Elon Musk announced a project called OpenAI along with a group of other people, notably Sam Altman, Greg Brockman, LinkedIn co-founder Reid Hoffman, Jessica Livingston, Peter Thiel along with Amazon Web Services, Infosys and YC Research, people initially took notice largely because of the high-profile Tesla chief’s involvement. In the years that followed, OpenAI has come to be synonymous with AI for the layperson.

Set up in 2015, OpenAI operated largely under the radar for the first five years of its existence as far as the general public was concerned. It did create a large enough buzz within tech circles to attract early investments from companies as varied as tech giant Microsoft and Indian IT services firm Infosys. OpenAI’s mission statement has been to create a safe and ethical path towards AGI.

In 2018, the organisation introduced the world to a generative pre-trained transformer (GPT)—a neural network designed to function like a human brain and trained on large data sets to provide relevant responses. Interest in what OpenAI was doing was still largely confined to technology firms. It was only in 2021, when it announced the launch of Dall-E, an image generation engine, that people outside tech circles started hearing about it and paying attention. Dall-E is an AI-based image generation engine that produces visuals based on text inputs provided by users. What made it unique is that the inputs were provided in natural language, which meant that the engine was accessible to anyone, not just a software engineer with coding capabilities. A user could ask it to create an image of ‘18th century Paris but with skyscrapers’ using exactly those words, and it would provide an image that matched these specifications. This was a game changer as it opened up the world of tech development to people with no tech or coding skills. An updated version, Dall-E 2, was launched the next year, and a few months later, OpenAI made its chatbot ChatGPT available to the world on 30 November 2022.

Within the first two months, the program had 100 million users, making it the fastest growing consumer application ever. This is when the non-tech world sat up and took notice of this Silicon Valley start-up, which until then had largely been unheard of outside Valley circles. The internet was abuzz with people sharing what prompts they had put into the app and what it threw back at times, often with unintentionally hilarious consequences. Some were amazed at its ability to provide code to carry out simple functions based on requests entered in conversational English. Others were mildly horrified that it stumbled over simple word problems that a first grader should have been able to solve.

One way or another, GenAI and the potential it had to simplify our quest for information had caught the world’s collective imagination. Enterprises, too, sat up and took notice. Tech firms wasted no time in announcing their intent to integrate GPT into their existing offerings.

Unsurprisingly, perhaps, Microsoft, which is a significant investor in OpenAI, was the first off the block. The company announced that ChatGPT would be integrated into its search engine and browser, with ChatGPT-generated responses being marked out from the regular search results to prevent any confusion for the users.31 Meta introduced Llama, its own open-source LLM, which has since been integrated into its offerings like WhatsApp.32 At almost the same time as Microsoft announced this collaboration with OpenAI, rival Google announced the launch of Bard, its GenAI chatbot that was later renamed Gemini. Bard was built on Google’s Language Model for Dialogue Applications (LaMDA), which it had launched two years earlier. Bard was launched as an experimental conversational AI offering powered by LaMDA and was largely seen as Google’s response to ChatGPT. In 2025, Google said that it was offering an AI mode on its Search function.33

These companies and their AI offerings have captured and dominated the wider public consciousness because of their massive user bases. However, the credit for rolling out the first GenAI system that was accessible outside a closed user group goes to Stability AI. While Google, OpenAI and others were working on closed-group versions of their GenAI products, Stability AI was the first to make it available to the larger population with the launch of Stable Diffusion, an image-generating tool, in 2022. The company believed that more people using AI tools would help make them better, at a faster pace.34

Hype or Game-Changing Tech?

Even in an industry that has come to be known for its hype cycles, the buzz created by GenAI has been unprecedented. It’s telling that as recently as mid-2022, there was almost no awareness around GenAI, which then skyrocketed to the top of all emerging technologies and ‘technologies to watch’ lists for the coming year, before companies started to get disillusioned by the lack of tangible results in 2023. This typically happens when everyone is keen to make the most of a new tech and jump onto the bandwagon—often without really evaluating whether they would actually benefit from it or not. Very often, the technology itself is at a very nascent stage, as in the case of GenAI, and not everyone can harness it to its full potential. Companies that get past this stage are likely to derive some long-term benefits from adopting the technology and will continue to work with it till it plateaus and becomes ingrained into their way of working. These companies need to parallelly work on how implementing AI-based tools will impact different aspects of their business, from people to processes. These leaders should ideally be spending an equal amount of time, if not more, on understanding the actual impact of this technology on their company and people and what steps they need to individually take to be able to manage this transition smoothly.35

While the hype around GenAI started only towards the end of 2022, AI-based tools were in the spotlight since 2020, largely due to the Covid-19 pandemic. The pandemic-induced global lockdowns accelerated the adoption and implementation of AI tools across industries and functions. Even vaccine development for Covid-19, which was done in record time, relied largely on these tools analysing huge swathes of data, something that hadn’t ever been done before.36 Manufacturing firms turned to AI to boost their supply chain effectiveness and rejig their production patterns as they worked with an additional shortcoming of global transportation systems being hit, impacting raw material availability. Some researchers have compared the AI wars to the space wars of the previous millennium, when nations were competing to see who would be the first to make it into space. With AI, too, they argue, it isn’t very different as countries explore how these tools can be used to improve military prowess, alongside manufacturing and other economic capabilities. Experts believe that countries will start making advances in areas where they already have an advantage or a superior position by integrating AI into them.37

For India then, there is a lot of focus on how AI can be used to improve its digital and software services. Multiple government departments are working on pilot projects to integrate AI into their offerings under the Government of India’s AI Mission.38 The potential impact of this scale of AI adoption, especially GenAI, may be likened to the smartphone revolution in India.39 One reason perhaps is that not since the advent of cheaper smartphones has any tech really had the potential to have an impact at as large a scale as this one, especially in India. Access to affordable smartphones and cheap, even free internet, ensured its rapid adoption even in smaller towns and villages, connecting people irrespective of their location and backgrounds. It also opened up access to goods and services as well as information and entertainment in a way that was previously unimaginable.

What’s striking this time around is that leaders are now attributing the need for fewer workers to the adoption of new technologies like GenAI.40 In the past, automation had not had a significant impact on jobs outside the manufacturing sector. As leaders grapple with this new, ever-changing world, they also need to focus on how to manage employee and investor concerns around the rapidly evolving impact of this technology. Microsoft chief Satya Nadella has repeatedly drawn parallels between GenAI and other major tech shifts in the last few decades—search, mobile and cloud.41 Even then, the potential AI can have on industries or on individuals is likely to far outstrip the impact of any of these. While the advent of search primarily changed the lives of individual internet users and cloud had an impact on enterprises, GenAI has the potential to significantly impact individuals, enterprises and governments alike. Sundar Pichai, Google chief executive, has gone on to say that GenAI is the biggest technology shift in our lifetimes and could be bigger than the internet itself. ‘It’s a fundamental rewiring of technology and an incredible accelerant of human ingenuity,’ he wrote in a blogpost on the search giant’s 25th anniversary.42 At the same time, the likelihood of this technology being misused and creating ethical and moral dilemmas far outweighs anything companies or individuals have had to grapple with in the past.

Deepfake proliferation has already resulted in scams across the board and presents a never-seen-before potential to create social unrest. Social media platforms and aggregators have ramped up the size of their fact-checking teams to try and minimise the impact and reach of these doctored or AI-generated videos and images.43 People who have worked closely on AI development over the last few years, decades even, are split down the middle on whether AI will overpower and overshadow humans. Geoffrey Hinton, who also won the Nobel Prize for Physics in 2024, has often been called a godfather in the field of AI, given the decades of AI research under his belt. He said that he is worried about the increasingly powerful machines’ ability to outperform human beings in ways that are not in the best interests of humanity.44 However, not everyone is as worried. Some experts suggest that it is important to have these guard rails and safeguards built into the very foundational layers of the technology to ensure that it doesn’t run rogue.45 Going ahead, it is perhaps important to remember that AI is just another technology, albeit one that has the potential to significantly alter our world unlike anything seen before. What we see happening around us at present is a re-imagining of the way we work and how businesses operate. Pharma, for example, is one industry where there is likely to be a large disruption, especially in the drug discovery process. Though still in the early stages, GenAI can bring down the drug design and discovery process, as well as predicting likely outcomes and side effects, to a few months instead of the few years that it normally takes. This was one reason why companies like Pfizer could quickly design and launch a Covid-19 vaccine in a record period of time.46 This can also change how products are designed and developed, especially electronic components in devices. Industrial design is another area where GenAI is already being used fairly extensively, as is chip design and material design.47 We already see some early implementations in areas like customer service, with the use of advanced chatbots that have almost completely eliminated the need for human interaction. In the future, the customer service industry will be further transformed, with the potential to make a customer service agent much more productive through the use of AI tools.48 While companies are currently experimenting with using AI tools to generate code, it’s quite likely that the AI tool will be able to create a whole new app or payment ecosystem based on a well-worded prompt. More recently, companies have started adopting agentic AI, replacing repetitive software jobs like quality assurance and testing with AI. This is expected to have a significant impact on entry-level jobs and has set alarm bells ringing for employees.49 At the same time, companies like Klarna, which had shifted almost their entire customer service function to AI, are now going back to hiring people after seeing significantly higher levels of dissatisfaction among customers.50 While there is no doubt that the volume of work being done by AI will only increase going forward, it remains to be seen exactly how this will impact jobs in the long run. Experts caution that while changes in how we work are inevitable, adoption of GenAI is unlikely to result in mass lay-offs or massive waves of specific jobs or roles becoming obsolete. What is more likely over the next few years is that white-collar professionals especially will start to use GenAI tools to augment their work. The more mundane parts of the job will be replaced by AI and automation tools, leaving the humans free to do the more complex aspects of the job. As with any new technology, while some roles become obsolete, AI is also creating new jobs, prompt engineers being the foremost one. Indian-language AI models are likely to power a lot of the new development in this field in India, which again will result in increased demand for trained professionals. For leaders, this is an important time to ensure that their existing workforce is suitably equipped to deal with the changes that are coming in the next few years, even months. Given that this is an evolving skill, it may not be possible to directly hire from the market or from competition to fill the job roles. A better approach would be to invest in training the existing workforce. Identifying high-potential employees and providing further specialised training to them is one way of ensuring that the company doesn’t fall behind in the AI race. Irrespective of the industry, it’s still early days when it comes to AI adoption. Most organisations are using open-source models and existing data sets to run pilot projects, but this is likely to change in the next few years. Once organisations start working with their own proprietary data, they are likely to drive more synergies from using AI, and this is the shift that will truly impact jobs going forward. More roles will be done with greater augmentation from AI tools, which will result in a fundamental shift in how leaders need to view jobs—both the ones that will be obliterated by AI and the new ones that will come into play. It’s not hard to imagine a situation five years down the line when the entire organisational structure will have to be re-imagined and rethought of to incorporate the tech changes that have taken place. Also, outside of pure play technology companies, other industries, too, are ramping up their hiring of tech talent. There is no doubt, though, that the job and labour market and the workplace as we know it will change significantly in the next five years.

In a conversation with Microsoft founder Bill Gates in 2024, OpenAI CEO Sam Altman said that the AI systems we have currently are set up to do tasks, not jobs. Over the next few years, these will evolve, and people will be able to do more with them. People will be able to use their data and connect it to outside sources to use it more gainfully. ‘The speed with which society has to adapt and how the labour market will change is the scary part (with GenAI) compared to other technologies,’ said Altman.51 It is expected that over the next five to ten years, AI systems will be fully equipped to deliver precise results, often with minimal prompts. Software developers are already working with AI tools to speed up the software development process, and over time, coding copilots are expected to become the norm.

Another study of over 2,700 AI researchers done at the University of California, Berkeley, has found that with the widespread adoption of GenAI, there is a possibility that all human tasks will become highly automatable by 2116.52 While this is still a while away, this prediction has been moved up by half a century from the original estimate of 2164, which was made in 2022. More immediately, these changes are likely to start reflecting in how tech and non-tech corporate jobs evolve in the next decade. This will have implications for not only how companies are run but also how people skill and train themselves. For aspiring CEOs who are in senior or mid-management roles at present, understanding how these changes will shape the future of their industries and organisations could be crucial when making a bid for the top job. Understanding how the technology landscape is evolving and what it means practically is a skill that will help set potential leaders apart from other contenders.

Despite the rapid progress that’s been made in the last few years, it’s useful to step back and take stock of what this is likely to be leading up to. Companies are just about starting to introduce multimodal AI capabilities with mixed results. Multimodal AI is a machine learning model that can process information across different formats or modalities, such as text, images and videos. In real terms, this means that a user can upload a picture of a chocolate cake and have the AI generate a recipe for it—or vice versa. While GenAI is a larger, umbrella term for different types of machine learning models that can be used to create new content from a prompt, multimodal AI capabilities expand the possibilities, and users are no longer restricted to only one type of input/output.