Find out why companies should adopt AI solutions as a service.
For every application available on-premises, it is almost certain that it will also be available as a cloud-based service, delivered on demand, by a cloud service provider. AI as a Service (AIaaS) is a fairly recent addition to the burgeoning field of cloud-based services. With AIaaS, companies can reap the benefits of AI without having to make upfront investments in hardware and software. And in the case of AI, the savings can be significant.
After decades as fodder for science fiction movies, the use of artificial intelligence in business has exploded. Businesses are using AI for everything from customer service and marketing to process automation, security, and business and commercial forecasting. In fact, a study conducted by strategic advisors NewVantage found that nine of the top 10 companies continuously invest in AI. A 2019 study by computer science researcher Gartner found 37% of organizations in 2019 actually used AI in the workplace.
SEE: Artificial Intelligence Ethics Policy (TechRepublic Premium)
However, for small and medium businesses, according to the same Gartner report, only 29% said they had adopted AI. This is at least somewhat influenced by the knowledge that specialized AI hardware is needed and is often prohibitively expensive. Indeed, a generic off-the-shelf server could be used, but due to the massive processing power required, it is not ideal and would kill productivity.
And that’s just the investment for the hardware. Then there’s the software, programming, and model training, which requires specially trained data scientists, who earn significant salaries. With AIaaS, businesses of all sizes can reap the benefits of AI research, machine learning, and analytics on demand and through the cloud.
When should enterprises adopt AIaaS?
Like all other technologies, AI was adopted slowly and incrementally. Companies are dipping their toes in the water before diving in headfirst to try it out and see if it delivers what it promises. Thus, the initial deployment of the first AI projects is generally measured and modest. Small businesses are particularly risk averse.
AIaaS is especially useful for companies that don’t expect to do a lot of AI work upfront. AI is broken down into a two-step process: training and inference. The training part is the computationally intensive part, but the inference has much lower power requirements and can be handled with a much less powerful unspecialized processor.
Now suppose you only plan to deploy two or three AI projects and have chosen to invest in specialized hardware. Since you cannot reuse an AI training server as a general purpose database server, it will remain unused.
Conversely, if you do multiple AI projects each year, you might consider taking a hybrid approach and investing in an on-premises system. This is because cloud services use a pay-as-you-go model for all the computing power needed to ingest and process the data along with all the associated applications for storage, databases, networking, and computing. to analyse. Ambitious AI projects generate massive amounts of data. Known as “data gravity,” AIaaS projects can multiply the need for additional capacity and services, driving up costs. This can easily blow up the cloud service provider (CSP) bill, and eventually it becomes more economically feasible to bring these workloads on-premises.
How AIaaS democratizes AI
There are a variety of programming languages for AI, from the common and ubiquitous (Python, C++) to the esoteric (R, Rust). This can be difficult for a non-data scientist, who may not have coding ability or understanding of data science beyond the basics. And too often, non-data scientists are tasked with owning AI projects because there simply aren’t enough skilled programmers and data scientists to meet the ever-increasing demands for their skills.
Fortunately, CSPs that offer AIaaS services also offer no-code frameworks for non-programmers. No-code tools and services are those that allow users to build applications without having to program them in the traditional way of writing, testing, and debugging source code. Instead, core functionality is created using visual tools much like a flowchart, where actions are taken based on predefined conditions. If you already use Microsoft Visio, you know how it works.
The absence of code allows business users to do the work of programmers, but the downside is that apps tend to be simplistic. If you want precise and precise control and action of complex AI models, you still need to program the application.
But no code is very good yet to start writing simple AI applications, ease the burden on data scientists who have much more demanding tasks ahead of them, and maybe write a simple chat bot.
Finally, the pros and cons of an AIaaS approach or an on-premises/hybrid approach to AI should be carefully considered and consider cost, time, and labor specialization. For those starting or completing a limited number of AI projects each year, the benefits of AIaaS can far outweigh the alternatives.
Phil Brotherton is vice president of solutions and alliances at NetApp.