Artificial Intelligence (AI) is everywhere. However, AI is often discussed in theory or in a futuristic manner which can make it difficult to understand. So, how can AI be applied in a practical way to help your organization innovate right now? In this post (and podcast) we’ll dive in and demystify AI to help you better understand how it can be applied to your business.
Stay up to date on what's happening in the digital transformation space by signing up for our newsletter, here.
What is AI?
Artificial Intelligence has now become the catch-all phrase for any application or product that is somewhat smart or intelligent. There is a finer distinction between AI, Machine Learning, and Deep Learning, but we’ll stick with the vague definition for now. Our goal here is to discuss the role of AI for organizations and their roadmaps. Let’s get started!
Is AI the Right Fit For My Organization?
Virtually every industry can benefit from leveraging AI, from mining, to e-commerce, to creative studios and healthcare. The problem of turning data into actionable insights exists across all industries. Fortunately, the approach that Data Scientists and ML Engineers take is widely applicable, independent of the data and insights of a specific industry.
For example, AI has made a significant impact on E-Commerce. Businesses with many online users need to know what products or services to offer and when to offer them. Many successful businesses have turned to recommendation engines and reinforcement learning to inform what products or services should be shown to a user and when to do so in order to create the higher conversions. Creating personalized and relevant user recommendations can help significantly in higher overall brand engagement and measurably better revenue.
Another notable area is the potential for AI in healthcare, which cannot be underestimated. While the benefits are seemingly endless, some promising areas include automated diagnosis, triage and referral, while also leveraging the health data of an entire population to identify the most effective treatments.
While the areas mentioned hold a vast amount of potential, one industry that has actually been at the forefront of adopting AI is the creative industry. Generative Adversarial Networks can create highly realistic images, video, text, designs, music. These models consist of two networks that try to outsmart each other. For example, the first network tries to create a realistic looking portrait (as pictured below). The second network then tries to correctly identify these forgeries in a stack of pictures that include real ones as well. Both networks push each other to get better and better until the forgeries become indistinguishable from real portraits.
Image produced by a GAN (generative adversarial network) via thispersondoesnotexist.com
These techniques allow for creative studios to augment their production with AI assisting in the creation of novel new content in powerful new ways. These are just a few examples, but it should be clear that AI does not have any inherent industry bias. It can be applied to augment and automate many tedious, repetitive, laborious, and even creative tasks.
How Does My Organization Get Started With AI?
Many organizations have barely scratched the surface in getting started with AI despite all of the media hype. While there is excitement around the possibilities, there are also questions regarding the costs and ultimate value to the business.
So our answer to the question of how to get started? Small.
The best way to start with AI is the same way to start with any other solution - don’t start with the solution. Start with the problem. AI is really just another tool in your toolbox. Granted, a very powerful one, but solutions to non-existing problems don’t generate value, whereas identifying the right problem can ensure the creation of an effective solution.
Working with cross-functional teams ensures that AI is viewed realistically and approached from a product development perspective with an focus on showing value quickly. After all, nothing promotes the necessary changes for organization-scale AI transformation more than successful AI applications in production.