Research shows that nearly half of new business artificial intelligence projects are abandoned midway.
A recent survey by international law firm DLA Piper of 600 key executives and decision makers from global companies revealed the significant challenges that businesses face in integrating AI technology. AI has the potential to revolutionize many sectors, but there are many obstacles to its successful implementation. This article explores these challenges in detail and offers expert commentary on how to navigate the complex landscape of AI integration.
The survey revealed that more than 40% of organizations are concerned that their core business models will become obsolete if they do not adopt AI technologies, while almost half (48%) of companies that have embarked on AI projects have had to pause or abandon them. The main reasons for these setbacks include data privacy concerns (48%), issues around data ownership and inadequate regulatory frameworks (37%), customer fears (35%), the emergence of new technologies (33%) and employee concerns (29%).
“The survey results are not surprising, if not surprising,” said Aric Feingold, president and chairman of Commit, a technology company that advises large enterprises on adopting AI-powered tools and provides AI-based tech solutions. “Based on our industry knowledge, we estimate that the number of companies that have started to consider adopting AI tools but ultimately decided to hold off for the time being is probably greater than 50 percent.”
But in contrast to the survey findings, Faingold believes the reasons why organizations abandon AI projects are fundamentally different.
“One of the main reasons is that there is a gap between the capabilities of AI-powered tools at their current stage of development and the processes these organizations are trying to streamline, some of which cannot yet be adequately addressed by tools available on the market,” he said. “This gap is relatively easy to identify even early in the process.”
“Another reason the move to AI-powered tools has stalled is the difficulty of integrating multiple disciplines, including data, cybersecurity and user interface,” he added. “This is something we at Commit have done regularly for many years in other contexts, but those without experience in this field may face difficulties.”
Feingold explained that customer service and support are the areas where AI is currently delivering the most gratifying improvements and efficiencies.
“Many organizations are already deploying chatbots and using AI to respond to customers more efficiently and quickly,” he said. “But the picture is still weak when it comes to software development tools, and as a result, adoption is also low. We expect this gap to close significantly in the coming months and years.”
Orna Kleinman, managing director of the SAP R&D Center in Israel, emphasized that in the world of Business AI, where the stakes are significantly higher, responsible, relevant and trustworthy data is paramount.
“The consequences of bias, errors, or 'hallucinations' within business AI models could have devastating consequences for companies, leading to lost revenue, damaged reputations, and even impacts on society itself,” Kleinman warned. “For businesses to trust generative AI, they need to be confident that their data is being handled responsibly and securely, and that relevant data is being taken into account. By design, generative AI tools must respect and adhere to data privacy, data ownership, and data access restrictions, and only operate in areas where explicit consent has been obtained.”
The three R's: relevance, reliability and responsibility
Kleinman emphasized that the three “R's” — relevance, reliability and accountability — are the foundations of trustworthy AI for the business world.
Based on the survey findings and expert insights, Kleinman pointed to several strategies that emerge for companies looking to successfully integrate AI.
“It's important to have a clear AI strategy that outlines the vision, goals and specific use cases with clear KPIs,” she says. “This strategy should be integrated into a broader business plan to ensure alignment and consistency. Investing in data governance is equally important. Establishing a robust data governance framework can help address concerns around privacy and ownership, including implementing clear policies around data collection, storage and use, and ensuring compliance with relevant regulations.”
Kleinman and Feingold emphasized the importance of fostering collaboration between different departments within an organization: They say cross-functional teams offer diverse perspectives and expertise that can lead to more innovative and effective solutions.
“In addition to strategic alignment and data governance, selecting the right AI vendor is key. Companies need to navigate the complex landscape of technology providers and select a partner that can meet their specific needs,” Faingold added.
According to Feingold, “Google, AWS and Microsoft platforms are well equipped to address data privacy concerns.”
“Companies should be aware that privacy concerns in the context of AI are not so scary and should not deter them from considering this technology,” he added. “Cloud providers have the skills to manage these concerns. This also applies to regulation, which is evolving and changing, but leaves plenty of room for companies to operate without taking on unnecessary risk.”
As they navigate the complexities of AI integration, businesses face numerous challenges and decisions. From aligning AI initiatives with strategic goals to fostering a culture of innovation, successful implementation requires careful planning and collaboration.
As Kleinman and Feingold highlighted, the stakes are particularly high in the field of business AI, where a misstep could have negative effects on revenue, reputations, and even society at large. As companies continue to grapple with these complex issues, one thing is clear: the path to AI adoption must be guided by a commitment to transparency, responsibility, and ethical practices.
