Explainer: Deep Dive into AI and the Sustainability Question
According to the Lawrence Livermore National Laboratory, supported by the US Department of Energy, the idea of Artificial Intelligence (AI) was first introduced in the 1950s by the English mathematician and computing pioneer Alan Turing.
Since then, AI has expanded and evolved into numerous approaches, each offering distinct abilities and alternative methods.
So, what exactly is AI?
The history of AI
According to a timeline published by IBM, the groundwork for AI was established in the 1950s, when AI was first defined as “human intelligence exhibited by machines,” building on the vision of Alan Turing.
By the 1980s, AI had evolved into machine learning, marking the era when AI systems began using historical data and expanding the databases behind them.
In 1997, IBM’s Deep Blue achieved a defining milestone by defeating world chess champion Garry Kasparov, a huge milestone for AI, according to Amazon Web Services (AWS).
The 2010s ushered in deep learning, as AI models started to simulate the functions of the human brain and move away from rigid, rule-based approaches.
Today, AI models are defined by Gen AI, applying deep learning techniques to produce original, human-like content.
“In 2022, chatbots that use AI and natural language processing (NLP) to have human-like conversations and complete tasks like OpenAI's ChatGPT became widely known for its conversational abilities, further boosting AI interest and development,” AWS says.
What is AI? According to Google Cloud
AI is a set of technologies that enables computers to carry out complex tasks such as image recognition, language translation, data analysis and personalised recommendations.
It forms the backbone of much of today’s digital innovation, driving value for individuals and enterprises alike.
One clear example is optical character recognition (OCR), which uses AI to extract text and information from documents and images, transforming unstructured content into structured data that delivers actionable insights.
At its core, AI is the discipline of creating systems that can reason, learn and act in ways that typically demand human intelligence, or process information at a scale far beyond human capacity.
It is a wide-ranging field that draws on disciplines such as:
- Computer science
- Data analytics
- Statistics
- Engineering
- Linguistics
- Neuroscience
- Philosophy
- Psychology
In business, AI is implemented through technologies primarily driven by machine learning and deep learning.
These capabilities are applied to tasks such as data analysis, prediction and forecasting, object recognition, natural language processing, personalised recommendations and intelligent information retrieval.
Google Cloud explains how AI works
Although AI techniques vary, they are all rooted in the common foundation of data.
By processing vast datasets, AI systems are able to learn, adapt and identify patterns or correlations that might elude human observation.
Central to this process are algorithms – structured sets of rules or instructions that determine how data is analysed and how outcomes are decided.
In machine learning, one of AI’s most prominent branches, these algorithms are trained on labelled or unlabelled data to classify information or produce predictive insights.
“The future of AI is not about replacing humans, it’s about augmenting human capabilities,” explains Sundar Pichai, CEO of Google.
Deep learning builds on this approach by employing multi-layered artificial neural networks designed to mirror the structure of the human brain.
Through continuous training and refinement, these systems enhance their performance over time, powering applications such as image recognition, language translation and a wide range of other advanced capabilities.
What are the negative impacts of AI?
According to Earth.Org, AI is often praised as a transformative technology, but its environmental impact is becoming a growing concern.
The development, operation and disposal of AI systems generate a significant carbon footprint, primarily due to the immense energy required to train increasingly complex models.
The Information and Communications Technology (ICT) sector’s emissions are projected to reach 14% of global totals by 2040.
Studies indicate that training large AI models can produce as much carbon emissions as hundreds of transcontinental flights, while the rising volume of electronic waste poses additional risks from toxic materials polluting ecosystems.
Beyond emissions and electronic waste, AI-driven technologies like autonomous vehicles, drones and intensive agricultural tools contribute to increased resource consumption, biodiversity loss and raise complex ethical concerns.
Earth.Org highlights a critical lack of transparency in AI development processes, which restricts accountability and obscures a clear understanding of AI’s full environmental and societal impact.
"Addressing these challenges requires energy-efficient hardware and algorithms, responsible design standards, ethical disposal practices and stronger regulation, supported by collaboration between businesses, researchers and policymakers to ensure innovation progresses without compromising environmental sustainability," Earth.Org explains.
Can AI be sustainable?
According to Intel, AI holds significant promise as a powerful driver of sustainability when applied responsibly.
Leading companies are already leveraging AI to reduce their environmental footprint, advance decarbonisation efforts and identify new opportunities for efficiency and growth.
Just as IT was pivotal to digital transformation, AI and technologists now play a central role in sustainable transformation, helping organisations cut the carbon impact of their IT operations and harness technology to achieve net zero.
From optimising supply chains and energy grids to enabling precision agriculture and smarter data centres, AI accelerates progress toward environmental, social and governance (ESG) goals while delivering new business value and long-term resilience.
