Generative Artificial Intelligence (AI) has been heralded as the next big thing in technology. Its advocates promise to revolutionize industries by automating tasks, generating content, and enhancing decision-making processes.
Word from – Dr Errol Brandt, founder and chief innovation officer, Knowledge Orchestrator
Yet, despite the hype, the results have been disappointing. It’s an open secret in the tech world that up to 85% of generative AI projects fail—more than twice the failure rate of typical projects. According to Gartner, a leading technology advisory firm, nearly a third of corporate AI projects are expected to be abandoned by the end of 2025.
Researchers cite several reasons for these failures, including poor data quality, a lack of relevant data, and insufficient understanding of AI’s capabilities and requirements. While these factors are certainly significant, the real challenges lie within the system itself, especially within the Australian manufacturing sector, which has been lagging in technology adoption for years.
The Unique Challenges of Australian Manufacturing
Manufacturing in Australia has been under immense pressure since the 1980s. Globalisation, rising energy prices, and a highly concentrated retail sector have squeezed profit margins, forcing manufacturers to operate on increasingly tight budgets. This financial strain leads to leaner Information Technology departments and diminished appetite for ambitious, “blue sky” projects.
Unlike banks and telecommunications companies, manufacturers often lack the profit margins needed to support long-term investments in technology and innovation. As a result, many internal IT projects face lengthy backlogs, with essential activities like cybersecurity, system upgrades, and disaster recovery consuming vast amounts of resources and time.
Specialist resources—those with deep process knowledge and expertise—are fully dedicated to these business-critical activities, leaving little opportunity to explore and adopt new technologies like generative AI.
The Skills Gap and Steep Learning Curve
Another significant hurdle is the skills gap within IT departments, as the underlying technologies used by generative AI are not core competencies for most IT teams.
The technologies that underpin generative AI—such as Linux and Python—are not commonly used in many manufacturing companies, which often rely on a traditional Windows and SQL technology stack that functions in entirely different ways. This results in a steep learning curve for IT professionals who lack the time to reskill and stay current with the latest updates.
Larger companies, on the other hand, often invest in data science teams capable of creating innovative machine learning models. These models are valuable in principle, but incorporating them into business processes is a separate challenge. Developing a solid technology solution is one thing; getting people to change their behaviors is another.
This disconnect between development and deployment is the real reason behind the high failure rate of AI projects.
The Necessity of Cultural and Process Change
Generative AI has the potential to reinvent entire processes, but it’s not as simple as rolling out a new software tool. Successful implementation requires significant changes in processes, which in turn demand adjustments across people, processes, and culture. These areas are not typically core strengths of internal IT departments.
Without buy-in from all levels of the organisation, even the best, most innovative technologies will fail to deliver their promised benefits. In many cases, the human aspect is far more challenging than the technology itself.
Overcoming the Barriers: A Holistic Approach
To overcome these challenges, organisations must adopt a more holistic approach to implementing generative AI. Here are some steps that can help:
- Invest in Training and Skill Development: Equip your IT staff with the necessary skills to work with AI technologies. This includes training in programming languages like Python and familiarizing them with Linux systems. Encouraging continuous learning helps bridge the skills gap and keeps them updated on the latest advancements. Even if they don’t become proficient, they will at least recognize that the mindset required is different.
- Foster a Culture of Innovation: Encourage collaboration between departments and promote an organisational culture that embraces change and innovation. Avoid the trap of thinking that Artificial Intelligence belongs solely to the IT department. Projects driven by the business, with support from IT, have a much greater chance of success.
- Engage with External Expertise: Consider partnering with AI startups or consultants who can provide the expertise and support needed to navigate the complexities of AI implementation. External partners can offer fresh perspectives and specialised skills that may not be available in-house. Don’t waste valuable time reinventing the wheel when startups have already solved many of the fundamental technical challenges.
Conclusion
Generative AI is transformational, but unlocking its potential requires more than just technical capability. It demands a strategic approach that addresses all aspects of change. The challenge is even greater in the Australian manufacturing sector due to technological constraints and outdated skills.
By acknowledging the deeper issues at play and taking proactive steps to address them, organisations can increase their chances of success with generative AI projects. It’s time to move beyond the hype and tackle the real challenges head-on, ensuring that the transformative power of AI doesn’t become just another missed opportunity.
Generative AI, when implemented thoughtfully and strategically, can provide significant competitive advantages. The journey may be challenging, but the potential rewards for those who navigate it successfully are substantial.
