The effective deployment of AI in the enterprise requires a focus on achieving business goals. Rushing towards an “AI strategy” and hiring someone with technical skills in AI to lead the charge might seem in tune with the current trends, but it ignores the reality that innovation initiatives only succeed when there is a solid understanding of actual business problems and goals. For AI to work in the enterprise, the goals of the enterprise must be the driving force.
Every serious technology company now has an Artificial Intelligence team in place. These companies are investing millions into intelligent systems for situation assessment, prediction analysis, learning-based recognition systems, conversational interfaces, and recommendation engines. Companies such as Google, Facebook, and Amazon aren’t just employing AI, but have made it a central part of their core intellectual property.
As the market has matured, AI is beginning to move into enterprises that will use it but not develop it on their own. They see intelligent systems as solutions for sales, logistics, manufacturing, and business intelligence challenges. They hope AI can improve productivity, automate existing process, provide predictive analysis, and extract meaning from massive data sets. For them, AI is a competitive advantage, but not part of their core product. For these companies, investment in AI may help solve real business problems but will not become part of customer facing products. Pepsi, Wal-Mart and McDonalds might be interested in AI to help with marketing, logistics or even flipping burgers but that doesn’t mean that we should expect to see intelligent sodas, snow shovels, or Big Macs showing up anytime soon.
Sponsored by Accenture
How it will impact business, industry, and society.
As with earlier technologies, we are now hearing advice about “AI strategies” and how companies should hire Chief AI Officers. In much the same way that the rise of Big Data led to the Data Scientist craze, the argument is that every organization now needs to hire a C-Level officer who will drive the company’s AI strategy.
I am here to ask you not to do this. Really, don’t do this.
It’s not that I doubt AI’s usefulness. I have spent my entire professional life working in the field. Far from being a skeptic, I am a rabid true believer.
However, I also believe that the effective deployment of AI in the enterprise requires a focus on achieving business goals. Rushing towards an “AI strategy” and hiring someone with technical skills in AI to lead the charge might seem in tune with the current trends, but it ignores the reality that innovation initiatives only succeed when there is a solid understanding of actual business problems and goals. For AI to work in the enterprise, the goals of the enterprise must be the driving force.
This is not what you’ll get if you hire a Chief AI Officer. The very nature of the role aims at bringing the hammer of AI to the nails of whatever problems are laying around. This well-educated, well-paid, and highly motivated individual will comb your organization looking for places to apply AI technologies, effectively making the goal to use AI rather than to solve real problems.
This is not to say that you don’t need people who understand AI technologies. Of course you do. But understanding the technologies and understanding what they can do for your enterprise strategically are completely different. And hiring a Chief of AI is no substitute for effective communication between the people in your organization with technical chops and those with strategic savvy.
One alternative to hiring a Chief AI Officer is start with the problems. Move consideration of AI solutions into the hands of the people who are addressing the problems directly. If these people are equipped with a framework for thinking about when AI solutions might be applicable, they can suggest where those solutions are actually applicable. Fortunately, the framework for this flows directly from the nature of the technologies themselves. We have already seen where AI works and where its application might be premature.
The question comes down to data and the task.
For example, highly structured data found in conventional databases with well-understood schemata tend to support traditional, highly analytical machine learning approaches. If you have 10 years of transactional data, then you should use machine learning to find correlations between customer demographics and products.
In cases where you have high volume, low feature data sets (such as images or audio), deep learning technologies are most applicable. So a deep learning approach that uses equipment sounds to anticipate failures on your factory floor might make sense.
If all you have is text, the technologies of data extraction, sentiment analysis and Watson-like approaches to evidence-based reasoning will be useful. Automating intelligent advice based on HR best practice manuals could fit into this model.
And if you have data that is used to support reporting on the status or performance of your business, then natural language generation is the best option. It makes no sense to have an analyst’s valuable time dedicated to analyzing and summarizing all your sales data when you can have perfectly readable English language reports automatically generated by a machine and delivered by email.
If decision-makers throughout your organization understand this, they can look at the business problems they have and the data they’re collecting and recognize the types of cognitive technologies that might be most applicable.
The point here is simple. AI isn’t magic. Specific technologies provide specific functions and have specific data requirements. Understanding them does not require that you hire a wizard or unicorn to deal with them. It does not require a Chief of AI. It requires teams that know how to communicate the reality of business problems with those who understand the details of technical solutions.
The AI technologies of today are astoundingly powerful. As they enter the enterprise, they will change everything. If we focus on applying them to solve real, pervasive problems, we will build a new kind of man-machine partnership that empowers us all to work at the top of our game and realize our greatest potential.