A report from Accenture has revealed that only a few companies (16%) have managed to make artificial intelligence (AI) work at scale. Meanwhile, the majority of the companies are still stuck in dead-end pilots.
The report titled “AI: Built to Scale”, shows how difficult it is for companies to transform as by a successful AI implementation. The report was derived after surveying 1,500 C-level executives across 16 industries to learn about what makes AI projects successful.
Bob Berkey, MD, Accenture Applied Intelligence stated in the report that several companies are struggling to scale AI and are stuck conducting AI experiments and pilots. However, they are achieving a low scaling success rate as well as a marginal return on their AI investments. As per the report, getting the AI transition right has substantial financial returns. It stated that there are three phases in the AI evolution:
Efficient use of data
The report found that most successful enterprises invest heavily in data governance frameworks, data quality, and data management. It revealed that companies that scale AI are likely to wield a larger and more accurate data set. Moreover, these firms are integrating internal and external data sets as a standard practice. As per the report, they make use of the right tools such as data and analytics search, data engineering, and data science workbenches.
Focus on building teams
While implementing AI, companies should first make sure the work is in line with company leadership. Next, it is vital to form a diverse team to manage the allotted work. Companies need to have a team that comprises data scientists, visualization experts, machine learning, data, and AI engineers. Moreover, they need to conduct ongoing training to ensure employees have an understanding of how AI applies to their role and take care of their implementation.
Focus on the investment
The report revealed that the companies that have invested significant time and effort in scaling AI, set longer timelines for AI projects. The company heads are more realistic when it comes to scaling the project what it takes to do so responsibly.
Moreover, companies need to have flexible business processes and clearly defined accountability. Finally, the report stated that the size of the company does not matter in the case of AI application. With the help of an equipped team and the right mindset, any company –small or big is likely to succeed.
Sneha Bokil is a Senior Editor with OnDot Media. She writes editorials on an array of topics ranging from IoT, AI, ML, and cloud computing, among others. She has over 9 years of experience in the field of content creation, where she has written on technology, both enterprise and consumer, and finance.
