Surprisingly Useful AI Article Enhancements
Sustainability has been firmly established as a clear strategic priority for Australian businesses. More than eight out of ten (85%) place high importance on achieving their environmental goals. To accelerate progress, a growing number of businesses are adopting artificial intelligence (AI) to optimise operations, reduce energy consumption, and cut emissions. However, despite growing adoption of AI, it remains a challenge to translate its potential into measurable, long-term outcomes. Many organisations are yet to look beyond AI’s monitoring and reporting functionalities, leaving an untapped opportunity to move from insight to action.
The recent Global Sustainability Barometer Study by Kyndryl and Microsoft underscores this disconnect between intent and impact. It revealed that despite a willingness and acknowledgement of AI-driven sustainability, challenges like data integration across multiple systems and regulatory constraints are holding businesses back from fully integrating and deriving the impact.
The untapped potential of AI
AI adoption in sustainability initiatives is gaining momentum, with 75% of Australian organisations using AI to monitor energy usage and emissions. While this marks a promising start, monitoring alone does not equate to meaningful change. AI’s value lies in advancing from observation to prediction—where it can drive tangible outcomes such as optimising energy management, forecasting demand, and identifying inefficiencies. These capabilities enable organisations to take proactive steps in reducing their carbon footprint.
In conversations with our customers, it’s clear that businesses are increasingly seeking to use AI to improve energy management, from integrating distributed energy resources to predictive maintenance. To truly advance sustainability efforts, AI must shift from a passive monitoring tool to an active enabler—guiding real-time decisions and driving smarter, more efficient operations that reduce environmental impact.
Data integration hinders AI adoption for sustainability
One of the biggest challenges Australian businesses face in adopting AI for sustainability is integrating data across multiple, disparate systems. Kyndryl’s study found that over half (58%) of local organisations identified this as their top challenge. This is followed by the barriers of regulatory compliance (55%) and lack of technology expertise (47%). These challenges not only slow down AI adoption but also prevent organisations from scaling their sustainability efforts effectively, limiting the broader potential AI can have on reducing emissions and optimising resource use.
Without a strong, unified data foundation, AI’s capacity to generate actionable insights is significantly limited. Whilst the complexity does make it difficult for businesses to scale their AI efforts, overcoming it means prioritising data consolidation so AI can move to driving real, measurable impact across energy management, emissions reduction and new growth opportunities.
This involves adopting data fabrics and vector databases that support advanced AI tools like large language models, enabling more accurate and scalable insights. Strong data governance is also critical—ensuring data quality, integrity, and compliance through practices like cataloguing, classification, and policy enforcement. By aligning sustainability goals with broader technology modernisation efforts and using open integration platforms that connect disparate systems in real time, organisations can shift from isolated data silos to a connected, insight-driven environment.
Environmental impact should be considered
While AI is increasingly recognised as a strategic tool for sustainability, many organisations are still struggling to measure its environmental impact and manage associated risks. Nearly two-thirds (62%) of Australian businesses currently don’t fully understand the environmental impact of their AI solutions, and over half (52%) report that AI has, in some cases, negatively affected their carbon emissions. Without clear visibility into AI’s footprint, businesses risk undermining their own sustainability goals, making it crucial to assess AI’s energy consumption and overall lifecycle impact.
To fully realise AI’s potential in advancing sustainability, it’s essential to embed responsible AI practices from the outset — accounting not only for the ethical implications but also the environmental impact. With AI increasingly underpinning sustainable innovation, businesses must take a holistic view of its influence and adopt responsible AI frameworks that govern both development and employment. Recognising these complexities, more than 28% of Australian organisations are now engaging technology partners like Kyndryl to co-develop clear roadmaps, establish governance structures and build the sustainable technology capabilities needed for lasting impact.
Australian businesses are at a pivotal moment — where AI has the potential to be a powerful accelerator of sustainability, but only if implemented strategically with purpose and responsibility. Moving from passive observation to action requires more than monitoring; it calls for integrated data, ethical governance, and a plan to use AI for measurable, impactful change. To truly leverage AI’s potential for sustainability, businesses must invest in robust data infrastructure, embrace responsible AI practices, and seek the right expertise to guide integration and transformation.
Done right, this approach can help position Australian enterprises as global leaders in ethical, tech-enabled sustainability.
Erandhi Mendis is Social Impact Lead at Kyndryl ANZ.
Last Updated on May 1, 2025 by Erandhi Mendis
