AI has caused quite the stir in the world of business , acting as a catalyst for transformation. Case in point: AI can crunch numbers at speed and scale. This means quicker and more informed decision-making besides the fact that AI can read a vast amount of data to create predictive analysis. Addo this, AI’s ability to recreate and better customer experience. AI has given businesses the ability to personalise interactions, anticipate needs, and provide solutions. Furthermore it has aided in lowering the cost of operations . Automation of routine tasks means companies can allocate resources more strategically, cutting down on operational costs.
But, and there’s always a ‘but,’ right? AI isn’t all rainbows and butterflies. Let’s not forget the ethical dimension as AI systems become more sophisticated, there are concerns about bias, privacy, and accountability. In conversation with Brandwagon Online, Frank Shaw, chief communications officer, Microsoft shared his views on impact and implementation of AI in businesses. (Edited Excerpts)
How is AI impacting businesses? When it impacts businesses somewhere, it also ends up impacting lives inadvertently. So how would you see AI impacting businesses and lives?
Basically, we’re in the early stages of a major shift in technology and so anybody can predict with any degree of certainty what is likely to happen and will probably be wrong.
This shift in technology will be kind of like when the Internet or mobile phones came in, first people use these products and it eventually transforms businesses. Right now, people are using AI to make things easier or more enjoyable, like having tools to transcribe meetings or help with computer commands. Industries will transform based on these possibilities. Just like we rely on tools like Copilot and Teams to answer our questions and transcribe meetings. Copilot and Windows help us to communicate more effectively with our computers, enhancing our capabilities. These improvements in existing technologies lay the groundwork for the evolution of industries.
For example, in healthcare, better transcription can improve patient experience, and AI can assist healthcare providers with up-to-date information. As this evolves, we might see more changes like increased telemedicine or AI-based screenings. Education is another area where AI can reshape how we learn, offering personalised assistance and sparking discussions about the best learning methods. It’s a major platform shift, and the full impact is yet to unfold.
If we talk about AI and marketing, focusing on the context of Google’s move towards third-party cookie diminishment, Microsoft has been proactive in this area way before by giving the option to switch on or off the cookies Was there any impact of cookies being disabled? How is the marketing stack? And how much has AI powered that now?
I think that the changes in which cookies work are still playing out. There are certain regulations in different regions. Everybody is adapting to work with the same but it is still unclear exactly is it going to be useful or not? For some individuals, the perception of cookies as detrimental stems from the notion that these digital trackers essentially mimic memory without their awareness. This lack of transparency raises concerns, leading to scepticism about the overall improvement brought about by these mechanisms. On the contrary, there are others who view the evolving cookie landscape positively. Their optimism is rooted in the belief that change empowers them with increased control over their privacy.
The marketing tech stack holds much potential for increased personalisation in engaging with potential customers. Previously constrained by scale, where a single person managing multiple customers struggled to deliver personalised experiences, the integration of large language models transforms this dynamic. This shift allows salespersons to concentrate on their core strengths, ensuring superior outcomes for customers. The resulting personalisation feels tailored to individual objectives, departing from the ineffectiveness of one-size-fits-all solutions. This evolution signifies a departure from traditional approaches, embracing a more individualised and effective strategy in the marketing technology.
Does the transition towards digital transformation, particularly in medium-sized businesses in the Indian market, mirror trends observed in the US?
I believe newer companies have a slight advantage since they haven’t made as many data-related mistakes as established ones. The challenge with AI in businesses today is its reliance on a coherent data structure. For long-standing companies, scattered data poses a complication in integration. On the other hand, newer companies can start with best practices, making AI adoption smoother. These firms may not need to invest as much in capital expenditure. If a company is already committed to the cloud, leveraging large language models becomes accessible. The key is to evaluate whether the adoption yields improved productivity, increased profits, and enhanced research efficiency. Our role is to guide businesses in assessing these outcomes.
What has been the most significant challenge you encountered with the implementation of AI?
I believe the major obstacle lies in the perception and understanding of AI itself. The way it’s often portrayed as a singular, mysterious force—a black box, a job-snatcher, or a world-changer. However, I find it more intriguing to view AI as a toolkit, offering various tools that individuals can utilise in unique ways. For instance, if you’re a writer and someone says, ‘I can write,’ it might not resonate well. But if they say, ‘I have a research assistant that you can use whenever you need,’ it becomes an appealing prospect. Shifting the perspective from AI as a threat to a useful tool is a challenge.
Generative AI or Explainable AI: which vision has a long way to go?
I think making predictions is tough. We’re just scratching the surface of generative AI, and the potential is really high. Some of the issues we’re facing, things that don’t work well, will be solved faster than people think. So, there’s a lot of potential.