Activities
Divisions
Programs
Activities
Divisions
Programs
FP Special: Amit Luthra, Managing Director (ISG), Lenovo, elaborates on the requirement of Autonomous Data for Independent AI
Roughly 93% of corporations in India aim to put resources into AI in 2024, and they are considering not only GenAI and LLMS, but also AI Inferencing models. In this situation, the creation of Sovereign AI is essential, but it will be extraordinarily challenging without Autonomous Data.
Despite the remarkable progress of artificial intelligence, we've only scratched the surface. Our attention has mainly been on generative AI and LLMs, but there's a lot more happening in different areas and it's advancing at a much quicker pace.
For instance, 93% of Indian corporations intend to pour money into AI by 2024. Their interests aren't limited to GenAI and LLMS, but expand to AI inferencing models that aid in understanding and scrutinizing scenarios, and forecasting necessary actions. Even though GenAI has sparked both fascination and fear, for Indian enterprises, the future lies in AI analytics, automation, edge-computing, and AI inference.
We had a conversation with Amit Luthra, the Director of Infrastructure Solutions Group at Lenovo India, during the launch of their 'CIO Playbook 2024 – It's all about Smarter AI' report. We discussed the future of AI in India.
We cover everything from the concept of data autonomy to the moral structures surrounding AI, from the misconception of AI "taking" our employment to the ways in which CIOs intend to utilize AI, not merely for its hype, but for its ability to create significant change. Here are the selected parts from our dialogue:
What changes has India observed in its computational requirements over time, particularly in relation to AI? India's digital evolution was based on four key elements – Social Media, Mobility, Analytics, and Cloud Computing. In the field of Analytics, there was always an element of AI. Most business software has attempted to incorporate AI. However, two significant events brought about a shift in this landscape.
Initially, the emergence of IoT led to a massive surge in data produced by sensors. Businesses recognized that they couldn't afford to accumulate the data and then analyze it retrospectively to achieve their objectives. For instance, if a manufacturing plant desired to enhance their production quality, they needed to scrutinize their data concurrently with its collection, particularly if the data was time-specific. This amplified the demand for Edge Computing.
The next step involves the introduction of GenAI and the development of large LLMs. These require advanced systems with significant computational capacity to construct large LLMs.
In my opinion, although a significant portion of current investments in India are concentrated on the development of LLMs, we will need to shift our attention towards enhancing investments in AI inferencing in the coming times. The growth rate of inferencing is expected to be much higher. This would involve a combination of CPU-GPU workloads, rather than solely depending on GPU-based systems.
What are the primary obstacles for today's businesses, particularly startups, when implementing AI? Particularly, generative AI requires a well-organized structure. The paramount issue is establishing efficient data management and maintaining data integrity. Feeding your AI models with poor quality data will result in poor quality outputs. Even if you develop a superior LLM that can learn rapidly and utilizes impressive algorithms, if the training data is incorrect, incomplete, outdated, or of poor quality, it will eventually cause problems.
It's equally crucial to establish a structure for analysis that aligns with distinct objectives. Businesses must have a clear understanding of why they require an AI system and what they intend to accomplish with it. They should question, "What is my goal with the data I've collected?"
Many organizations' AI initiatives fall short due to a lack of clarity about their intended use of AI. While some companies might aim to enhance customer experience with their AI projects, others may be seeking to boost their production levels.
Some might desire to establish cleanliness procedures to enhance efficiency, while others might just aim to deter theft and scrutinize fraudulent activities. Companies must identify their ultimate goal and understand what they are aiming to realize. Once that's established, the technology will naturally align.
There are two critical aspects to consider – the aspect of security and the aspect of ethics, both demanding equal attention. You are handling data, which can either elevate you or pull you down. Companies must question how to tackle the ethical side of it and determine boundaries. Moreover, they must make sure to manage sensitive data correctly, surrounded by suitable safety measures and security structures.
The CIO Playbook 2024 notes that nearly half of all CIOs face hurdles in filling AI-focused positions. Despite India boasting the most substantial group of engineers and developers, there seems to be a gap between the industry's needs and the existing capabilities of these professionals. AI has been part of the engineering curriculum for roughly 25 years. However, applying what's taught in engineering institutions to the modern business environment is a different ballgame. It's unrealistic to assume that a recent graduate can immediately demonstrate proficiency in AI work.
The issue isn't a lack of resources, but rather establishing a supportive structure for these resources.
This presents a significant difficulty as situations evolve rapidly. What was deemed critical in the previous year might be irrelevant this year. The query a business proprietor may be pondering now is how can they consistently guide and educate their staff, and how can they constantly facilitate a learning environment?
Next, we encounter the problem of retention. The act of stealing employees, or poaching, is a significant concern, particularly in the context of Artificial Intelligence. In this situation, companies must consistently put resources into their assets and be determined to keep them.
There's a considerable amount of discussion about Autonomous Sovereign AI and Data Sovereignty at the moment. Given the current progress of LLMs, what obstacles do you foresee in fostering sovereignty? Data sovereignty involves understanding the location of your data and making sure it stays within your country's geographical borders. Typically, data might be kept on the cloud, but the actual hosting of this cloud could be in a different nation.
For instance, numerous multinational corporations previously had unified data centres situated anywhere worldwide, aiming for enhanced and fluid operations. However, in such scenarios, data sovereignty is impossible since the data, which may include personal details of Indian nationals, is stored beyond the country's borders.
In a situation like this, the importance of data sovereignty law cannot be overstated. It ensures that personal information of Indian citizens, for instance, remains within the country's borders.
After confirming this, the security and analytics structure along with the government enters the picture. These structures must guarantee that essential data is not exposed, and it is correctly managed and stored according to the rules established.
So, how do Indian companies utilize this information and construct an analytics structure based on it? It's crucial to understand your goal and what you're aiming to accomplish. Each sector and application varies.
In this context, what types of LLMs or apps might we expect to see created? Here's an illustration. If I need to contact the customer support team of a service I'm utilizing, I'd rather phone them than interact with a chatbot via an app, as the experience is often dissatisfying. However, we are now witnessing the introduction of AI-powered voice assistants in customer service roles, where all interactions are verbal.
A further illustration is the retail sector. In the near future, traditional salespeople may be replaced by digital avatars designed to address your unique enquiries, depending on your specific needs. Imagine walking into a showroom and requesting a salesperson to elucidate the attributes of a new vehicle. A less experienced salesperson might leave the customer perplexed. Furthermore, there's a possibility they might not be able to respond to your inquiries if you pose a very specific, yet unusual question related to the product.
Visualize the same situation, but with a virtual character that resembles a human being and who is fully knowledgeable about the product it represents. After only one training session, they will be capable of responding to all your inquiries.
In addition, the company doesn't have to allocate substantial time to retrain and refocus employees when updates occur. During this procedure, it's crucial to ensure that whether big or small, the language models are learning in real-time and have access to the most recent data sets. Think about how distinct earlier versions of ChatGPT are compared to the GPT4.5-powered ChatGPT.
After the construction of the LLM, it's crucial to ensure its portability, contemporariness, and most significantly, its seamless deployment. This is where Lenovo steps in, offering organizations a spectrum of solutions from compact to cloud-based, since AI isn't confined to just one LLM. AI actually spans across the whole ecosystem.
The Indian government hasn't yet implemented rules concerning AI within the country. From a business standpoint, what kind of stipulations would you prefer to see in these regulations? Furthermore, how might the Digital Personal Data Protection Act, established in 2023, affect the progress of AI in India? It's going to be critical to establish a robust Security and Ethics Framework. How can we protect individuals from deepfakes that are already in circulation? The emphasis will be on security and ethics in this conversation. Once the landscape is understood, companies that are creating LLMs will be able to identify boundaries.
Top executives and business heads, despite not necessarily possessing extensive knowledge in AI or technology as a whole, are enthralled by the idea of incorporating AI into their enterprises. CIOs are tasked with guiding these leaders, ensuring AI is properly integrated and applied in appropriate areas. Our research for the CIO Playbook 2024 demonstrated that the desires of a business and the requirements of technology often diverge. Chief executives are particularly interested in generative AI, likely due to its popularity as a trending term.
In 2023, Generative AI didn't exist, but now it's the most critical focus as it's a trending topic. Everyone was concerned with customer satisfaction and experience last year. But now, it's their second most important goal because everyone understands that attracting a new customer is costly, whereas keeping an existing one is far simpler.
Likewise, income and profitability expansion, which has always been the top concern for all companies, has now dropped to the third place. Sustainability, which was never really a concern, has now risen to be the fifth most important priority, as businesses aim to operate ethically.
Let's now examine the priorities of the CIO or the tech-oriented aspect of a business. CIOs still maintain that automation should be the top priority. By focusing on automation, many issues will be addressed. Therefore, AI, particularly generative AI, should be considered only after implementing automation.
For AI to effectively handle high-performance tasks, it requires a specifically designed infrastructure. Furthermore, the centrality of security and ethical considerations in your AI cannot be overstressed. As mentioned before, these components collectively form a framework. AI cannot function autonomously; it must serve a purpose. Introducing AI is akin to devising a strategy, complete with various pillars and underpinned by a framework. We have always stood by this belief, and it's gratifying to see it confirmed in this survey when we revisit and engage with CIOs.
What sectors should be prioritized by the government when incorporating AI into their administrative tasks? How can AI improve the effectiveness of government operations? The government has a strategic perspective on AI. As citizens who have voted for the government, it's important to align with their direction. When government officials and authoritative bodies engage with us, industry leaders, in forums, they gather various insights and opinions on a range of topics. They also encourage us to contribute our viewpoints and suggest optimal practices.
I believe that once the government establishes an AI strategy and introduces a structure, such as data sovereignty, all will need to adhere to it. Moreover, our Prime Minister, Shri Narendra Modi, has also spoken about 'AI for All'. Similarly, we at Lenovo have been advocating 'AI for All' for several years now. We also hold the conviction that when we achieve AI for all, the vision will be significantly more robust.
Ultimately, the burning question everyone wants answered is: How will AI transform the current employment landscape and the nature of jobs as we understand them? AI is simply another piece of technology or a collection of tools designed to enhance human performance. To illustrate, think about the wars we've fought – World War I, World War II, and every other conflict. During World War I, .303 rifles were the weapon of choice for soldiers. Currently, they are equipped with LMGs and MMGs. Despite the significant evolution in technology and tools, the need for soldiers remains.
AI is simply another resource similar to firearms, another facilitator that will allow them to perform more efficiently and manage advanced equipment.
Indeed, AI will transform the characteristics and possibly the caliber of jobs. However, it won't eliminate jobs or render them obsolete. While some types of roles might become unnecessary, it will also create new opportunities.
I'm of the view that AI will likely lead to an increase in wages, given the necessity for individuals to acquire new and enhanced skills. Self-improvement will become crucial to avoid becoming irrelevant. Consider the transformation in the IT industry. Two decades ago, during job interviews, I would frequently inquire if applicants were familiar with SAP sizing or exchange sizing. Today, such questions are no longer relevant and have fallen out of use.
When individuals note in their resumes that they have a strong command of PowerPoint in the current era, but lack understanding of integrating it with AI, their skills may no longer be applicable. AI is a tool that aids everyone in evolving. Ultimately, it's crucial to remain current, employable, and pertinent to the industry. To achieve this, personal investment is necessary.
Search for us on YouTube
Highlighted Programs
Associated Articles
NVIDIA's Jensen Huang believes AI hallucinations can be fixed and that artificial general intelligence is roughly 5 years away
Apple has at last introduced MM1, its artificial intelligence model for generating text and images
Microsoft recruits Mustafa Suleyman, co-founder of DeepMind, to head its new consumer AI division
Samsung and Rebellions, chip-making firms from South Korea, have plans to outdo NVIDIA
NVIDIA's Jensen Huang asserts AI hallucinations can be addressed and estimates artificial general intelligence to be about half a decade away
Apple has finally unveiled its multimodal AI model MM1 for text and image creation
Microsoft appoints DeepMind's co-founder Mustafa Suleyman to lead its fresh consumer AI department
Samsung and Rebellions, South Korean chip producers, aim to surpass NVIDIA
Can be found on YouTube
Firstpost holds all rights and protections under copyright law as of 2024
+ There are no comments
Add yours