Keynote-Olivier Blum

19 Feb 2026 14:45h - 15:00h

Session at a glance

Summary

This discussion features Olivier Blum, Global CEO of Schneider Electric, speaking at an AI conference about the intersection of artificial intelligence and energy infrastructure challenges. Blum begins by reflecting on his journey from managing Schneider’s India operations in 2007, when access to reliable power was a major issue, to now leading the global company as the world faces similar energy challenges on a planetary scale. He emphasizes that AI will dramatically increase energy demands, as more computing power requires significantly more energy consumption, creating unprecedented pressure on global energy systems.

Blum explains that data centers are evolving rapidly, with power requirements per rack increasing from a few kilowatts to potentially 1 megawatt in future AI applications, necessitating entirely new electrical architectures like 800-volt DC systems. He argues that while much attention has been paid to clean energy supply since the 2015 Paris Agreement, insufficient focus has been placed on making energy demand more efficient. According to Blum, the world will need an additional 10,000 terawatts of electricity between 2024 and 2035, with AI adding even more pressure beyond current electrification scenarios.

However, Blum sees tremendous opportunity in what he calls “energy intelligence” – connecting physical energy systems with digital technology and AI to optimize efficiency. He believes this approach can reduce energy consumption by 10-30% across applications, using the example of AI-managed home electrical panels. Blum concludes by highlighting India’s potential as a global innovation hub for energy-AI solutions, noting that Schneider Electric’s largest R&D center and workforce are located there, positioning the country to lead this technological transformation.

Keypoints

Major Discussion Points:

– AI’s massive energy demands and infrastructure challenges: The discussion highlights how AI will dramatically increase global power consumption, with data centers requiring exponentially more energy (from a few kilowatts per rack to potentially 1 megawatt), creating unprecedented pressure on energy systems worldwide.

– The evolution from energy access to energy efficiency: Blum traces his journey from addressing India’s basic power access issues in 2007 to now facing global challenges of reliable, clean power, emphasizing that energy efficiency on the demand side is as crucial as clean energy supply.

– “Energy Intelligence” through digital-physical convergence: Schneider Electric’s vision of connecting all energy systems digitally and applying AI/foundational models to create intelligent energy management, potentially saving 10-30% of energy consumption across applications.

– India’s strategic role in global energy innovation: India is positioned as a key hub for developing next-generation energy solutions due to its challenging operating conditions, cost-competitive environment, engineering talent pool, and innovation capacity.

– The urgency of reinventing energy infrastructure for the AI era: The need to develop new electrical architectures (like 800-volt DC systems) and completely reimagine energy systems to support AI’s growth while maintaining sustainability goals.

Overall Purpose:

The discussion aims to highlight the critical intersection between AI advancement and energy infrastructure, positioning Schneider Electric’s “Energy Intelligence” solutions as essential for managing AI’s massive power demands while achieving global sustainability goals. Blum advocates for India’s central role in developing these innovations.

Overall Tone:

The tone is consistently optimistic and forward-looking, with Blum expressing excitement about technological possibilities while acknowledging serious challenges. He maintains an enthusiastic, visionary perspective throughout, balancing realism about energy demands with confidence in technological solutions. The tone becomes particularly passionate when discussing India’s potential and Schneider’s personal connection to the country.

Speakers

– Moderator: Role/Title: Conference Moderator; Area of Expertise: Not mentioned

– Olivier Blum: Role/Title: Global CEO, Schneider Electric; Area of Expertise: Energy efficiency, digital infrastructure, power systems, AI-driven energy solutions

Additional speakers:

– Mr. Schneider: Role/Title: Not mentioned; Area of Expertise: Not mentioned (only referenced by the Moderator at the beginning)

– Prime Minister: Role/Title: Prime Minister (of India, based on context); Area of Expertise: Not mentioned (referenced by Olivier Blum in his remarks)

Full session report

This discussion features Olivier Blum, Global CEO of Schneider Electric, delivering a keynote address at an AI conference that positions energy infrastructure as the critical enabler for AI’s transformative potential. Blum’s presentation bridges his personal experience in India with global technological challenges, offering practical solutions for the AI-energy relationship.

Personal Journey and Global Context

Blum begins by congratulating the Prime Minister and Indian government for the event, then shares his personal connection to India. Reflecting on his appointment as Managing Director of Schneider Electric India in 2007 (landing in 2008), he recalls arriving in a country where access to reliable power was the paramount concern. Speaking almost 20 years later as Global CEO, he draws a compelling parallel: what was once India’s localized challenge has now become a global issue, amplified by AI technologies that demand unprecedented energy resources.

His six years in India proved foundational to understanding how innovation emerges from constraint. Blum notes that “if you can crack the code in India, we’ll crack the code everywhere,” reflecting his appreciation for India’s unique operating environment—characterized by equipment under extreme pressure, cost-competitive requirements, and exceptional engineering creativity, including what locals call “Jugaad.”

AI’s Exponential Energy Demands

Blum presents a detailed analysis of AI’s energy implications, emphasizing that “we have not even started the era of AI” while already facing significant power access issues globally, including peak load problems and power cuts even in developed nations like the United States.

The technical progression he outlines illustrates this challenge’s magnitude. Data centers evolved from requiring “a couple of kilowatts per racks” to 10, 20, then 30 kilowatts. Current projects in India demand approximately 80 kilowatts per rack, while cutting-edge facilities in the United States utilizing NVIDIA GPUs require 150 kilowatts. Looking ahead, Blum envisions future requirements reaching 500 kilowatts to 1 megawatt per rack, necessitating entirely new electrical architectures like 800-volt DC systems.

This exponential growth creates what Blum describes as “tremendous pressure on the energy system that forces every single company to reinvent the energy system.” Industry projections suggest over 200 gigawatts of new data center capacity will be needed by 2030, with 50% being AI-driven.

The Energy Planning Gap

Drawing from International Energy Agency data, Blum notes that supporting global economic growth will require an additional 10,000 terawatts of electricity between 2024 and 2035, followed by another 12,000 terawatts between 2035 and 2050. However, his research team at Schneider Electric believes these projections fail to account for AI’s additional energy demands.

This gap represents a critical blind spot in global energy planning. Blum observes that even at this AI-focused conference, insufficient attention was paid to energy requirements, reflecting a broader disconnect between technological ambitions and infrastructure realities.

Energy Intelligence: The Solution Through AI

Central to Blum’s presentation is “energy intelligence”—Schneider Electric’s approach to addressing AI’s energy challenges through AI itself. This represents what he describes as a historic first: connecting physical and digital worlds after 190 years in the power sector. The strategy involves making all energy assets connectable and applying AI foundational models to create intelligent energy management systems.

Blum argues that energy intelligence can achieve 10-30% energy consumption reductions across all applications. He provides a compelling example of home energy management, noting that residential electricity consumption represents the largest category of global electricity use. By connecting electrical panels in homes and applying “AI agents” for automated energy management, significant savings become possible. “I can prove it. I’m testing it in my own home,” he states, lending personal credibility to his claims.

India as the Global Innovation Laboratory

Throughout his presentation, Blum positions India as the primary laboratory for developing next-generation energy technologies that can scale globally. His rationale is compelling: India’s operating environment subjects equipment to extreme conditions, creating natural testing grounds for durability and performance, while cost-competitive requirements force innovation toward affordability and efficiency simultaneously.

India’s exceptional engineering talent and creative problem-solving capabilities make it ideal for breakthrough innovations. Schneider Electric’s commitment reflects this confidence—India hosts the company’s largest R&D center globally with 8,000 researchers and engineers, represents their third-largest market, and employs 40,000 people total, making it “the largest one in number of employees” for the company.

Blum’s partnership with NVIDIA has helped them understand next-generation chip requirements, further positioning the company to develop appropriate energy solutions for future AI infrastructure.

Technology Integration and Future Infrastructure

Blum’s vision encompasses fundamental changes in how energy infrastructure operates. The transition to intelligent, connected networks represents entering “a new era where all energy systems are connectable.” This connectivity enables advanced AI models to optimize energy flows in real-time, responding to demand patterns, supply variations, and system constraints with unprecedented sophistication.

The technical challenges are substantial. Developing 800-volt DC systems for next-generation data centers requires new equipment designs, safety protocols, installation procedures, and maintenance approaches. Creating truly intelligent energy management systems demands integration across power electronics, communications networks, data analytics, and AI algorithms.

Overcoming Adoption Challenges

Blum acknowledges that implementation has been slower than hoped over the past decade, with resistance from traditional operators and reluctance to adopt cloud-based systems for sensitive operations. However, he expresses optimism that AI’s transformative potential will overcome these barriers by demonstrating clear value propositions through cost savings, efficiency improvements, and operational benefits.

Moving Forward

Blum concludes by emphasizing the opportunity ahead. Rather than viewing AI purely as an energy consumer, he advocates recognizing its potential as an energy optimization tool. His message is clear: the AI revolution requires an energy revolution, and that energy revolution must be intelligent and efficient.

His emphasis on India’s role reflects broader themes about innovation emerging from constraint and the importance of diverse perspectives in solving global challenges. By positioning India as a potential leader in AI infrastructure development, Blum suggests that solutions to global problems may emerge from unexpected sources.

The presentation closes with Blum noting his plans to meet with government representatives and construction companies, indicating ongoing efforts to implement these energy intelligence solutions in India’s rapidly growing infrastructure.

Session transcript

Moderator

Thank you, Mr. Schneider, for your remarks. And ladies and gentlemen, I would like to welcome Mr. Olivier Blum, Global CEO, Schneider Electric. Schneider Electric sits at the intersection of two of the most pressing challenges of the AI era, energy efficiency and digital infrastructure. As data centers consume ever -growing share of global power, Olivier Blum is leading the company that is helping make that infrastructure sustainable. Please welcome the Global CEO of Schneider Electric, Mr. Olivier Blum.

Olivier Blum

Thank you very much. So, first of all, I’d like to congratulate the Prime Minister and the entire Indian government and all the associates for this beautiful event. This week has been tremendous. And that put together an ecosystem of stakeholders who are really moving technology to the next level. Now, I’d like to tell you a little bit about my own story. I landed in 2008. In 2007, in India, I was appointed at that point of time Managing Director of Schneider in India to develop Schneider Electric. And I discovered a country where the major issue was really access to power, access to reliable power. Now I am sitting in front of you, you know, I’m standing actually in front of you as a CEO of a global company, Schneider, almost 20 years later.

And guess what? What is the biggest problem of the planet? Access to reliable and clean power. So it’s not only the issue of India, it’s the issue of the worldwide planet. And we are just at the beginning of a new era because we are facing issues with access to power in many geographies. You heard many stories of problem of peak load in different geography, including the US where you have power cut. And we have not even started the era of AI. What AI changed to the world? AI means more compute, more compute means more energy. And we just don’t underestimate today. We don’t know exactly what is going to take, but that’s going to put the pressure on the energy system, which has nothing to do with what it is today.

So we know that energy is already a priority for government, for organizations everywhere in the world. It’s even a geopolitical topic. But we are entering a new era where AI could transform the planet. And since this morning, you have heard many, many speakers talking about how AI could impact our life, our businesses everywhere in the world. But the real question is, how we take the energy system to the next level and what AI can bring to make energy more efficient. So we do believe for a company like Schneider, it changed everything. Because we are the first time in the history of Schneider, after 190 years in the power sector, where we are at a point where we can connect the physical and the digital world.

And it was not possible before. Now, let me look back since 2015, what happened. In 2015, something very important happened in the world, which was the climate agreement, you know, the Paris Agreement, where… Well, I don’t know. The world has put a huge focus on energy. And if you remember at that point of time, a lot of people spent time on the supply, bringing clean energy everywhere in the world, which was great. But Schneider Electric was a very, very strong advocate to say, look, working on the supply is very important, but we have to spend even more time to work on the demand side, on how we make energy efficient. And guess what? In the past 10 years, companies like us, but not only, we’ve been really, really strong advocates that if we build a world which is more electric, if we build a world which is more electric and more digital, we might have a path really not only to decarbonize the planet, but to give access to energy everywhere in the world.

I think we’ve made good progress. We’ve made good progress, but it has been complicated because we faced a lot of resistance in the system. We faced resistance with poor grid actors. We faced resistance by companies implementing those new technologies. We know large companies don’t like necessarily to put all their data on the cloud. You know, they want to do it on purpose. Premise and so on. and so forth. But let’s make it clear, I don’t think all the technology we are ready everywhere in the world to make energy system much more efficient. So I think it has been a good journey in the sense that we’ve started really to electrify more and more in the world.

But I think we have just started to scratch the surface. Now, let me come to the topic of AI and the conference, and I will come back on energy. I’m sure you are aware that in any kind of report, we are talking about more than 200 gigawatts of capacity that needs to be built in data center in the next coming years, by 2030. We say usually that 50 % will be AI -driven, and we see the acceleration. For a company like Schneider, we see the acceleration since two years, and we started to see the acceleration when we built our partnership with NVIDIA, when we started really to understand the next generation of chips that will be used to NVIDIA.

For those who are not very familiar with what is a data center, we are talking, about a couple of years, about a couple of kilowatts per racks in the data center. Then we moved to 10, 20, 30. What we are building right now in India, it’s something which is already around 80 kilowatts per racks, even more. And what we are doing in the US right now with the GPU, which are available for NVIDIA, is already at 150. We are forcing a world with NVIDIA, where it can go to 500, to 1 megawatt. And that puts a tremendous pressure on the energy system that forces every single company to reinvent the energy system. I’m sure some of you have heard about the concept of 800 volt DCs, which are the new type of electrical architecture you will need to have for the data center of tomorrow that will be able to power the AI industry.

But again, I’m coming back. There is two phases of AI. There is one which is, there will be many, many new infrastructure that has to be built. But we have really to invent this next level of infrastructure to make sure that they can support what the AI would need. But the second part of the equation, which is even more exciting for me, is that probably we are entering in a new era where for the first time we can make energy more intelligent. But let’s be very realistic. There are a certain number of data from IEA which are telling that to support the global economy, the world will need another 10 ,000 terawatts of energy to be produced, of electricity to be produced between 2024 and 2035, and another 12 on top of that between 2035 and 2050.

When we are looking with our research team at Schneider Electric, we are building energy scenario. We believe that those scenarios don’t include what AI will bring to the planet. So the scenario in terms of electrification that you need for the planet is not about making usage more electrical and supporting the electrification of the planet for cars, for heat pumps, for electrification of process. What brings AI on top of that is another level of pressure on the energy system. And if you look at this conference, which has been great. We have learned a lot. We have spent a lot of time with great people. I’ve not heard enough about energy. And I’ve not heard enough about the need to make energy much more intelligent if you want to support the next AI journey.

Now, for a company like Schneider Electric, it’s really, really fascinating. We created the company 190 years ago. We have been a great company in the physical world. Ten years ago, and I mentioned the Paris Agreement, this is when we decided to make sure that every single asset that will connect, that will sell in the market, will have to be connectable. So we are entering a new era where all energy systems are connectable. And if we are able to apply all the great models, all the great foundational models, which have been built by a lot of partners with whom we are working, we can, for the first time in our history, connect the physical and the digital world.

And that’s what we call energy intelligence. And by doing so, we believe we can overcome many of the difficulties that we faced in the past 10 years to make energy systems more efficient. And by doing so, we believe we can overcome many of the difficulties that we faced in the past 10 years to make energy systems more efficient. That can eventually also solve one of the biggest problems of the planet, which is the climate transition. And by doing so, we believe we can overcome many of the difficulties that we faced in the past 10 years to make energy systems more efficient. because again it’s not only about clean energy it’s about more efficient and we believe we are entering in this new era where if we are able to connect system to collect data to apply foundational model that will connect the physical and the digital world we can save between 10 20 30 percent of energy consumption in every single application in the world and i’m just going to tell you and finish with one example think about your home you’re all sitting today in this conference or some of you are connected and while we are speaking there is a lot of energy consumption in your home tonight you will be back to your home you have an electrical panel somewhere in your home very likely your electrical panel is not connected today imagine a world tomorrow where every single connected if every single panel in the world electrical panel in a home will be connected imagine a world where you are able to extract data imagine a world where you can apply ai agent and where we can manage energy for you while you are not even in your home you can save again between 10, 20, 30 % of your energy consumption.

Actually, I can prove it. I’m testing it in my own home. So I’m telling you it works. Maybe you don’t know, but consumption of energy in home is the largest consumption of electricity in the world. So the good news, we are entering in a new era where we know that the world will be more electrical. And for the first time in our history, we can say that the world can become more electrical if we are able really to leverage the power of the new technology. And I will just finish by telling you that why I’m extremely pleased to be here in India, you know, almost 20 years after I started really my journey in this country, is because India has a lot of different factors which are very, very different than the rest of the world.

And what I’ve learned a lot by spending, you know, six years of my personal life in India, India is one of the countries where equipment is under tremendous pressure, more than in any other country. In the world. Point number two, India is an extremely cost -competitive country, where you need to bring the best innovation at the best price. Number three, the level of innovation. I know in India, some people say also Juga in India, but the level of creativity that you can have in India to create new systems that will solve the most complex problems of the planet can be done in this country. And when you look at the number of engineers you have in power, automation, digital, you have all the ingredients which are together at a point of time where we need to make AI a big transformation for the planet.

For a company like Schneider, I can tell you it’s not only words. India is the third largest country of Schneider Ethics in the world. This is the largest one in number of employees, 40 ,000 employees. You don’t know it, but this is the largest R &D center we have in the world with 8 ,000 employees. Largest number of software engineers. So I’m super excited because we are starting a new phase for the planet, a new technology revolution, which is called AI. where India can be the place where a lot of innovation starts with. I came to India, you know, 20 years ago where we are bringing a lot of product from outside. I think we are at a point of time where India can innovate the next technology that will make the world more efficient.

So we call that Schneider Electric Energy Intelligence. I’ve been very, very excited to be part of the summit. We’ve met a lot of people from government, from construction company, from technology company who have really this strong appetite to make sure that AI will bring progress for all and making sure that India can be at the center of that transformation. And I always say to my team, you know, because now I’m leading Schneider Electric globally, if you can crack the code in India, we’ll crack the code everywhere. Thank you very much.

Moderator

Thank you so much, Mr. Blum, for highlighting the technology. Thank you for highlighting all those facts which concern the power consumption. So far as AI is concerned,

Olivier Blum

Speech speed

192 words per minute

Speech length

India as a catalyst for innovation and Schneider’s growth

Topics

Capacity development | The enabling environment for digital development | Environmental impacts

Moderator

Speech speed

154 words per minute

Speech length

107 words

Speech time

41 seconds

Moderator’s framing of the discussion

Explanation

The moderator highlights that power‑consumption data is central to the AI conversation and frames Schneider Electric’s role at the intersection of energy efficiency and digital infrastructure.

Evidence

“Thank you for highlighting all those facts which concern the power consumption.” [4]. “Schneider Electric sits at the intersection of two of the most pressing challenges of the AI era, energy efficiency and digital infrastructure.” [11]. “As data centers consume ever‑growing share of global power, Olivier Blum is leading the company that is helping make that infrastructure sustainable.” [53].

Major discussion point

Moderator’s framing of the discussion

Topics

Artificial intelligence | Environmental impacts

Agreements

Agreement points

Recognition of AI’s significant power consumption challenges

Speakers

– Olivier Blum
– Moderator

Arguments

AI will dramatically increase energy consumption as more compute power requires more energy, creating unprecedented pressure on global energy systems

The moderator acknowledges the importance of power consumption facts related to AI technology development

Summary

Both speakers acknowledge that AI technology development creates significant power consumption challenges that need to be addressed

Topics

Artificial intelligence | Environmental impacts

Similar viewpoints

Both speakers recognize the critical importance of understanding and addressing the energy requirements and power consumption challenges associated with AI technology development

Speakers

– Olivier Blum
– Moderator

Arguments

AI will dramatically increase energy consumption as more compute power requires more energy, creating unprecedented pressure on global energy systems

The moderator acknowledges the importance of power consumption facts related to AI technology development

Topics

Artificial intelligence | Environmental impacts

Unexpected consensus

Limited discussion of energy requirements in AI conferences

Speakers

– Olivier Blum

Arguments

The conference highlighted AI applications but insufficient attention to energy requirements and the need for intelligent energy systems

Explanation

It is unexpected that a major AI conference would have insufficient focus on energy requirements, given that this is a fundamental constraint for AI development. This suggests a gap in how the AI community approaches infrastructure planning

Topics

Artificial intelligence | Environmental impacts

Overall assessment

Summary

There is clear agreement between the speakers on the fundamental challenge of AI’s energy consumption, with both recognizing this as a critical issue that requires attention

Consensus level

High consensus on the core issue of AI’s energy demands, though limited to only two speakers. The implications suggest that energy infrastructure planning must be integrated into AI development strategies, and that there may be a broader need for the AI community to focus more on energy requirements

Differences

Different viewpoints

Unexpected differences

Overall assessment

Summary

No disagreements identified as this was primarily a single-speaker presentation

Disagreement level

No disagreement present – this transcript represents a monologue by Olivier Blum with only brief moderator acknowledgments. The moderator’s single substantive comment actually reinforces and thanks Blum for his points about power consumption and AI technology, showing alignment rather than disagreement. To properly analyze disagreements, multiple speakers would need to present contrasting viewpoints or challenge each other’s arguments, which does not occur in this transcript.

Partial agreements

Similar viewpoints

Both speakers recognize the critical importance of understanding and addressing the energy requirements and power consumption challenges associated with AI technology development

Speakers

– Olivier Blum
– Moderator

Arguments

AI will dramatically increase energy consumption as more compute power requires more energy, creating unprecedented pressure on global energy systems

The moderator acknowledges the importance of power consumption facts related to AI technology development

Topics

Artificial intelligence | Environmental impacts

Takeaways

Key takeaways

AI will create unprecedented pressure on global energy systems, requiring 200+ gigawatts of new data center capacity by 2030 with power consumption per rack potentially reaching 1 megawatt

The world needs 10,000 terawatts of additional electricity between 2024-2035, plus another 12,000 terawatts by 2050, not accounting for AI’s additional energy demands

Energy intelligence through connecting physical and digital worlds can reduce energy consumption by 10-30% across all applications, representing a major efficiency opportunity

India is positioned as a strategic innovation hub with unique advantages including cost-competitive innovation requirements, engineering creativity, and Schneider Electric’s largest global R&D center

The energy transition requires focus on both supply-side clean energy and demand-side efficiency improvements, with AI enabling intelligent energy management for the first time

Current AI discussions insufficiently address the massive energy infrastructure requirements needed to support AI development and deployment

Resolutions and action items

Schneider Electric is implementing energy intelligence solutions by making all energy assets connectable and applying AI foundational models

The company is developing next-generation electrical architectures like 800 volt DC systems for AI-powered data centers

Schneider Electric is leveraging India as a primary innovation center to develop technologies that can be scaled globally

Olivier Blum is personally testing connected home energy management systems to prove the 10-30% energy savings potential

Unresolved issues

How to build the massive energy infrastructure required to support AI growth while maintaining sustainability goals

The gap between current energy planning scenarios and actual future AI energy demands

Resistance from grid actors and companies reluctant to adopt new digital energy technologies

The need for greater focus on energy requirements in AI development discussions and conferences

How to scale energy intelligence solutions globally across different regulatory and infrastructure environments

Suggested compromises

Impact

This comment elevates India’s role from AI consumer to potential AI infrastructure innovator. It suggests that solutions developed for India’s challenging environment could become the global standard, positioning the country as a leader rather than follower in AI infrastructure development.

Overall assessment

Explanation

Blum suggested India has unique factors (equipment pressure, cost competitiveness, innovation capability) that could position it to create breakthrough energy solutions, but didn’t specify what these innovations might be

Disclaimer: This is not an official session record. DiploAI generates these resources from audiovisual recordings, and they are presented as-is, including potential errors. Due to logistical challenges, such as discrepancies in audio/video or transcripts, names may be misspelled. We strive for accuracy to the best of our ability.