Findability Sciences

‘To Be or Not to Be’: AI Adoption in Traditional Enterprises

Ingesting AI as a piecemeal solution or a quick fix is not going to help traditional enterprises exploit the optimum potential of AI. A techno-cultural shift, driven by thought leadership at the top, leading to a broad-spectrum adoption of AI across multiple functions of an organization is the most critical paradigm shift that traditional enterprises need.

Emergence of AI in Enterprises

It would be appropriate to tag the last decade as the decade of resurgence of AI. Technically the AI winter lasted from 1974 – 1980 but even when the winter was over, for next 3 decades, developments in the field of AI were limited to confined lab experiments done by some key players like IBM which developed Deep Blue in 1997, a chess playing algorithm that beat the world champion Gary Kasparov. However, AI adoption for organizational decision making did not find its footing till the turn of the century. After being put to hibernation for almost three decades, AI was awakened from its cryogenic slumber.

Phenomenal development in computing speed, capability of parallel processing, partial, if not complete, democratization of data and technology paved the way for emergence of AI as a potent force that can usher businesses into the future-verse. The present decade started on a worrisome note of the pandemic which threatened to derail the global growth engine but, in the end, not only did human resilience trump the virus but the post pandemic era also witnessed accelerated adoption of technology and AI in day-to-day functioning of enterprises. Today AI seems to be a necessary spoke in the enterprise wheel and many leaders hailing from traditional world have also opined the same over last couple of years. Yet, all is not so smooth when it comes to welcoming this new member in a traditional family.

Challenges in AI Adoption in enterprises

1.The Culture Conundrum:

CXOs of traditional enterprises sometimes have been heard sharing their experiences about how AI ‘did not work’ for them. A deeper probe will highlight that most of such failures were driven by the lack of techno-culture that is needed to nourish an AI ecosystem rather than the failure of AI solutions themselves. Just as IT function has become an integral part of organizations over the past two decades, AI also needs to be considered as a necessary and integral business function that should be supported by focused thought leadership. This integration and cultural adoption will happen only if the CXOs move beyond looking only at the financial ROI coming from AI solutions and start giving equal importance to the capability ROI and strategic ROI that AI ecosystem yields. Not only do the business leaders need to adopt AI solutions, but also review business outcomes and performances around AI driven metrics. The urgency and necessity of data / machine empowered decision making must percolate from top to bottom and that is what will lead to a pragmatic and prudent adoption of AI solutions and a flourishing AI culture.

2. The Talent Turmoil:

The second biggest challenge traditional enterprises face is getting the right talent. The problem is not only limited to finding the talent, but it also stretches to the realm of deciding the team size and sustained engagement over a period. Given the lack of experience in managing AI functions, traditional enterprises struggle to assess and acquire the right talent. To a considerable extent, this can be managed by taking professional support from recruitment firms which specialize in executive sourcing in AI / ML space. The question of in-house team size must be evaluated cautiously. Traditional businesses do not need to have a mammoth Data Science team for multiple reasons -It might turn into a cost centre with suboptimal returns, inadequate number of diverse AI and automation projects for resource optimisation and engagement etc. A modest team with a fundamental understanding of the AI and technology can do the job of smoothly managing and coordinating with external specialists. There are numerous service providers and products available in the market, off the shelf, which can be leveraged without bearing the long-term cost of resources and making heavy infrastructure investments. One of the successful modes of creating scalable AI/ML teams is to use the BOT (Build, Operate and Transfer) model for creating AI COE in collaboration with AI organization of proven record.

3. The Cost Constraint:

Finally, the last, but not the least concerning challenge brings us to the question of commercials. It is imperative that just like expenses are fixed as ratio of revenue for functions like sales, marketing, and IT, similarly there must be a dedicated budget for Data Science and AI, and not consider it as a stepchild of the IT function. What budgetary commitments need to be made for AI function is a subjective decision, depending upon the industry sector the business belongs to and the extent of AI integration that needs to be done. However, what is necessary is that there must be a commitment of investment from the top leadership to create a sustainable AI adoption culture and environment.

Traditional enterprises should refrain from viewing AI as a magic potion or a quick fix solution. It must embed into the enterprise DNA which would require a combination of will, vision and wisdom to embark upon the journey of transforming a traditional enterprise into an AI empowered enterprise adapted to the future.

FAQs

1. When did the emergence of AI in enterprises happen?

From 1974-1980, only key players were experimenting in the space. IBM developed Deep Blue in 1997, a chess playing algorithm that beat the world champion Gary Kasparov. However, AI adoption for organizational decision making found its footing only by the end of the century.

2. What paved the way for AI in traditional enterprises?

3. What are the benefits of AI adoption in enterprises?

AI has emerged to be a necessary spoke in the enterprise wheel and many leaders hailing from traditional world have also opined the same over last couple of years. AI saves time and money by automating and optimising routine processes and tasks. It also increases productivity and operational efficiencies and makes faster business decisions based on outputs from cognitive technologies.

4. What are the main challenges in AI adoption for enterprises?

Lack of techno-culture that is needed to nourish an AI ecosystem is missing in most organizations. This integration and cultural adoption will happen only if the CXOs move beyond looking only at the financial ROI coming from AI solutions and start giving equal importance to the capability ROI and strategic ROI that AI ecosystem yields.

Just like expenses are fixed as ratio of revenue for functions like sales, marketing, and IT, similarly there must be a dedicated budget for Data Science and AI. What is necessary is that there must be a commitment of investment from the top leadership to create a sustainable AI adoption culture and environment.