AI-As-a-Service (AIaaS): What it is and What it Means for the Tech Industry
AI is big news. Whether you’re in the tech community or the general public, everyone is officially conscious of its existence and aware that it’s bringing big benefits to our businesses. Hooray. We’re finally at a place where everyone is on board, we all know we’ll be adopting AI at some point soon, and we even have great use cases of it being put into place right now.
What we want to do here is not another “why you need AI” conversation. What’s more curious, in our minds, is how this adoption will actually look moving forward.
Think about it for a moment. Developing bespoke AI technology is costly. It requires specialist resources and a hell of a lot of time to feed it information to get it resembling a remotely helpful tool. Most businesses don’t have the time, money or staff to develop such a thing.
Microsoft, Google, Amazon, IBM and the likes have all been working hard and investing big bucks in AI for some time. While many may have been concerned, in the beginning, that this would mean the tech giant superiors would eventually eat up all the smaller businesses, it actually means that the tech giant superiors are working on tools, much like the other “-as a service” type services, to help other businesses get access to AI without needing to start from scratch in its development.
AIaaS as it Currently Stands
Right now, what this looks like are broad stroke AI services that skilled teams can tap into to develop their own AI solutions for their businesses.
One example of this is Woolworths, who recently announced that they are trialling AI-enabled scales at a handful of locations in Sydney which can automatically recognise the type of fruit and vegetable being weighed. This type of solution creates much faster checkout experiences, thus better customer experiences, better sales, and so on.
That is now. But as the AIaaS industry grows, the offerings are going to become more niche, ushering in a new reality where everyone can tap into AIaaS that suits their particular needs. In the not too distant future, it is realistic that smaller supermarkets will have access to the specific tools for fruit and veg recognition via an AIaaS provider, without a Woolworths sized team to design it.
We already have:
- Bots and digital assistants
- Cognitive computing APIs
- Machine learning frameworks
- Fully-managed machine learning services
While all of these have significant impacts, the one that is going to be of most interest to those reading this article is the cognitive computing APIs. This allows developers to add specific technology without the need for coding, including NLP, computer speech, computer vision, emotion detection, knowledge mapping, translation, and search.
This is only going to advance.
AIaaS for the Future
We are heading toward a place where AI could build AI, and even sooner, we will have AI building apps and back-end systems. What this means for businesses and developers is an exponential jump forward in release time, meaning more work can get out to market faster.
While this, of course, brings up a whole conversation that could be had about what that means for jobs, the immediate impact is AI easing the flow of work; products being pushed out faster and built quicker, thanks to AI that can work while staff are asleep.