2025: The Year of AI — Beyond the Façade
Kanika Mishra Pathak · Posted on: February 4th 2025 · read
There’s no doubt that 2025 is the year of AI. Yet, public discourse about AI trends often skips over a few critical elements: the infrastructure that powers it, and the profound environmental implications of this technological foundation, as well as also whether its use in commercial applications is as near the top of business agendas as many commentators make out.
Nvidia, now recognised as the world’s 9th most valuable brand by Brand Finance (thanks to its decisive bet on AI), exemplifies how businesses have driven this revolution. Their innovations—from GPUs to high-performance computing and data storage technologies—form the backbone upon which the visible achievements of artificial intelligence and machine learning rest. These advancements are not merely predicted; they are essential for AI to thrive, permeate and become omnipresent in modern society.
But this progress comes at a significant cost. AI models, including the transformers that power the likes of ChatGPT, consume vast amounts of power. According to Bloomberg, this energy demand has, in some cases, been met by old-school coal plants, undoing progress toward achieving global ESG goals. This environmental toll underscores the urgency of addressing the unintended consequences of our AI-driven future, as evidenced by the climate-related disturbances that continue to escalate worldwide.
"An important development that has emerged amidst the growing concerns over AI’s environmental toll is the rise of Deepseek. This company made headlines when they managed to prove that powerful AI models could be run on less powerful chipsets."
This breakthrough significantly affected Nvidia’s stock prices, as Deepseek’s innovation demonstrated that AI models could function effectively with less energy-intensive hardware.
Even more significant is Deepseek's decision to open-source their code and research, marking a pivotal moment in AI development and paving the way for models that aren't bound by the resource-intensive hardware typically required for cutting-edge technology. As companies and developers build on these advances, we can expect a shift in engineering that eliminates the traditional energy and compute constraints of AI infrastructure. This new wave of accessible, less resource-hungry AI models promises to balance innovation with sustainability, driving a revolution toward more energy-efficient and environmentally conscious technologies in the future.
Another promising breakthrough lies in the integration of AI with quantum computing. Early research suggests that quantum transformers could consume significantly less energy than their classical counterparts, offering a path to mitigate the environmental impact of AI's rapid expansion. The convergence of AI and quantum computing might represent the next frontier in solving both the computational and environmental challenges that currently limit AI’s full potential. As these technologies evolve, the future of AI may not only be shaped by advancements in modelling and engineering, but also by a new wave of breakthroughs in sustainable, efficient computing - one that is both powerful and planet-friendly.
The Government’s AI roadmap highlights the role of AI to meet the challenges of achieving Net Zero carbon emissions. This includes work on improving access to data, establishing robust governance frameworks, and developing cleaner systems, products, and services. However, despite recent advances, there is a need to be pragmatic about AI's proliferation in everyday life. Concerns have emerged over the security vulnerabilities tied to the storage of Deepseek data on Chinese servers.
Technology
Read more about TechnologyRead moreThe convergence of AI and quantum computing might represent the next frontier in solving both the computational and environmental challenges that currently limit AI’s full potential.
While AI is widely acknowledged to reduce inefficiencies, there are concerns that some AI projects may be driven more by the desire to advance the technology itself, rather than addressing clear, practical needs. For many businesses we speak to whilst the future use and development of AI is inevitable, it is not a priority on their short-term agendas.
Data is gold in the AI revolution, and in the hands of hostile states, it can be weaponised. This introduces a critical need for heightened awareness, especially in sectors where security and reliability are paramount, such as Life Sciences. Beyond the environmental challenges and practical considerations, security considerations further underscore the importance of a balanced approach to AI’s growth.
The excitement around AI is justified, but as we move forward, it’s essential to remain mindful of the unintended consequences. 2025 could be a pivotal year for both AI’s proliferation and the conversation about how we can advance it responsibly - integrating both technological, environmental and ‘value for money’ considerations for a more sustainable future.
