Artificial intelligence (AI) is finding its way into more industries and several reports have shown that a growing number of companies are already experiencing the benefits of implementing AI. Despite the enthusiasm, a new IBM study shows that lack of AI strategy continues to thwart their AI progress.
The study shows, while a majority of IT professionals in India are exploring or deploying some form of AI tools and technologies in their organisations, a whopping 70% of respondents in the study said that ‘lack of AI strategy’ is the largest barrier companies face in developing explainable and trustworthy AI.
The study conducted by Morning Consult on behalf of IBM, said that more companies are recognising the value of AI as they emerged from the challenges of the Covid-19 pandemic and invested in their digital transformation. In fact, the study shows that AI adoption was up four percentage points compared with 2021. But talent and skills shortages continue to hamper AI success, as many are still finding out ways to integrate AI strategies with their business goals. Of the rest, only 28% mention having a holistic strategy in place, and 25% have a strategy that is focused only on limited or specific use cases.
Establishing trustworthy, responsible AI practices is yet another challenge, shows the study. While business leaders acknowledge that trustworthy AI is critical, majority of the organisations surveyed have not taken key steps to ensure their AI is trustworthy and responsible. The key challenges in this space reducing bias, tracking performance model drifts and making sure they can explain AI-powered decisions.
But a gap in responsible AI practices further show a lack of AI maturity within organisations.
The research also noted that smaller organisations are at a receiving end as they are significantly less likely to take advantage of AI. The survey determined top three barriers to AI adoption for those businesses that include limited AI expertise or knowledge (34%), high prices (29%), and lack of tools and platforms for developing AI models (25%).
IBM is not the only organisation talking about how lack of a solid AI strategy affects its success, in a September 2021 study, analyst firm Gartner mentioned that organisational unreadiness, lack of investment in trust and poor data quality can hinder AI implementation — a reason why businesses should come up with an AI strategy that needs to be integrated into its business objectives.
“AI adoption requires more than just the latest technologies or modelling techniques. When moving from a pilot AI solution into production — or scaling AI in the enterprise — IT executives need to articulate a clear business purpose and rationale to invest in these technologies,” the analyst firm said.
There is no such thing as a one-size-fits-all business case for AI, believe experts, and developing strategies for AI projects and training an AI model also require a long trial-and-error process. In this regard, Cem Dilmegani, founder of Estonia-based tech analyst firm AIMultiple, said in his blog, “Building a successful AI strategy also requires collaboration between data scientists, data engineers, IT professionals, designers, and line of business professionals an can never happen in silos.”