AI Adoption’s Hidden Costs Are Strenuous for Firms

AI Adoption’s Hidden Costs Are Strenuous for Firms

As companies push to integrate AI into their businesses on the back of its ability to boost productivity and efficiency, the same technology comes with hidden and strenuous costs.

A recent study has shown that companies are caught in between full adoption of the technology and the need to stay on track with their ESG goals as AI systems’ energy consumption skyrockets.

High energy costs

A report by data storage vendor Pure Storage in conjunction with Wakefield Research speaks of the challenges that companies face when trying to fully integrate AI systems into their operations. The report cites what the researchers term the “often overlooked energy requirements of this advanced technology.”

The researchers interviewed 500 IT buyers at companies with over 500 employees in the US and Europe. One of the challenges they cited is high energy consumption.

About 73% of the IT buyers indicated they were not prepared for the high energy consumption associated with AI systems. Their high energy consumption translates to the need for infrastructure upgrades, adding to the costs.

“Most organizations lack the necessary infrastructure to handle the high-performance data demands and energy requirements essential for maximizing AI’s benefits,” reads part of the report.

“And while AI brings immense promise, its impact on energy requirements can be surprising.”

According to the report, AI adoption is forcing businesses to upgrade their infrastructure. Data management tools (48%), data management processes (46%), and data storage infrastructure (46%) play a significant role in this regard, the report states.

Additionally, other infrastructure upgrades need to include computer infrastructure (43%), networking infrastructure (44%), and privacy tools or processes (44%).

The costs are not unique to companies alone, but reports show there has been a significant increase in both energy and water consumption at data centers, pushed by the increased adoption of AI.

For instance, chatbots are taking up energy much faster than renewable energy sources can supply, according to Forbes.

Also read: Standing Up Against AI is a Futile Fight, Says SAG’s Jason George

Huge setbacks

As a result, these challenges have set back companies’ sustainability goals. According to the study, 89% of the respondents are also finding it hard to meet their ESG goals because of updates to their IT infrastructure following AI adoption.

“These hidden costs of AI pose a challenge to the successful implementation of critical corporate initiatives, including those aimed at achieving environmental goals,” reads the report.

It further adds that as AI adoption increases, “IT teams require a meaningful data strategy to ensure they can efficiently and effectively operationalize AI through the right infrastructure.”

However, about 60% of those who have already adopted AI are in the process of upgrading their IT infrastructure or plan to invest in more energy-efficient hardware in the next 12 months to meet their ESG goals.

Pure Storage chief technology officer Rob Lee said: “Planning for change and ensuring flexibility are key to navigating AI adoption.”

“As power and data demands increase exponentially in the age of AI, investing in and deploying the right AI-ready data infrastructure is not only essential to effective deployment and energy efficiency but also to driving the most value out of AI projects,” added Lee.

Costs versus adoption

A separate survey by PwC in August this year showed that while businesses are encouraged to embrace AI, there is little guidance on the investments required to make the most of the fast-growing technology.

About 90% of the executives surveyed indicated they were also struggling to measure the return on investment in AI. Indications are that businesses are responding to pressures from external markets as user demand keeps growing.

While technology comes with its complexities and fears over job losses, businesses are still upbeat about it and willing to take the risk. Nearly half of the 509 executives who participated in the PwC survey said they would invest in AI in the next 12 to 18 months.

A Forbes report says AI is not for technology executives alone anymore but has become “just as much a job for business executives.”

Image credits: Shutterstock, CC images, Midjourney, Unsplash.