Artificial intelligence (AI) seems to have more hype surrounding it than any other technology right now. Businesses are excited to adopt AI solutions to help them better analyse their data, automate mundane yet critical processes and ultimately provide customers with differentiated experiences. But there’s not a one-size-fits-all approach. Every organisation will have different requirements linked to their strategic goals – and their budgets.
AI services and solutions can currently be separated into two broad sections. There is consumable, API-driven technology that is perfect for “horizontally scalable” workloads – that is, challenges that are the same regardless of the industry the company operates in. For example, translating different languages or converting speech to text or vice versa. These requirements are easily transferred across businesses and best solved by API-driven AI solutions such as IBM Watson.
Then there is a deeper level of technology that uses your own data and information to produce insights. And this can operate alongside API-based solutions to maximise your AI results. This deeper layer, which we offer as a holistic hardware and software solution called IBM PowerAI, provides businesses with the most popular machine learning frameworks to enable the rapid deployment of high-performance AI, that leverages your own data to provide insights you could not uncover before.
AI at work
So, this all sounds great – but how would these two AI services work together to benefit businesses? Let’s look at the example of a chatbot. Customer service is a vital part of any business, but it is time-intensive. AI can help you automate that process while providing even higher levels of service to customers. Businesses can leverage Watson’s APIs to process a customer’s query or complaint, whether that’s delivered verbally or in written format, while PowerAI can pull in and process an organisation’s data and use it to provide accurate, relevant and valuable solutions.
Another powerful example comes from retail. Inventory management is another laborious, though crucial, task. And for a small flower shop, it’s even harder. Flowers are counted by hand; each bloom is categorized and logged. AI can help. IBM Watson, can recognise whether a photo contains a flower/s, therefore filtering out irrelevant images. A retailer could then use PowerAI to classify these photos against its own digitised inventory, saving significant amounts of time to focus on other tasks.
The journey to AI is not without challenges. And making a business case is a fundamental requirement. It’s important to build a strategy that will deliver business value – but won’t take years to implement. There are typically three places to start when identifying a use case. The first is to look at any area of your business where you collect, store and process data. Could that process be made easier? Could you be doing more with that information? Secondly, talk to your staff, your boots on the ground as it were, and find out what their pain points are. What takes up a lot of their time? What would free them up to make a more material impact on your business. Their feedback will be invaluable. Finally, think about how you can get ahead of your customers’ needs. How could you pre-emptively provide value to delight current customers and reduce churn?
Another, crucial, challenge to overcome is determining the data you need to realise your AI ambitions. What information do you have in-house? Do you need to pull in data from outside sources such as social media feeds or weather forecasts? Many CIOs and organisations try to ascertain what data they need before they carve out a clear use case. Once you know what you want to do and achieve with AI, sourcing the right information to make it happen is easier.
There is a lot of hype around AI – and that excitement is justified. By understanding what services are on offer, how they work together and what you want to achieve with them, you can not only adopt AI successfully to benefit your customers – you can use it to gain a competitive advantage.
AI is taking off – are you ready?
IDC expects 25% of all workloads to be AI, in 24 months time. Read how you can get ready in this IDC White Paper – Hitting the Wall with Server Infrastructure for Artificial Intelligence.