A recent newsletter from Peter Diamandis’ Abundance Insider inspired the writing of this article. If you are an entrepreneur and you aren’t familiar with Peter, I highly recommend you check him out. You have likely heard of some of his ventures like the XPRIZE and Zero-G. You can subscribe to his newsletter, Abundance Insider, here.
There seems to be some general confusion around terms like SaaS, PaaS, and newer iterations like MLaaS and AIaaS. Admittedly, for a long time, I didn’t realize there was much of a distinction and grouped all of these under the SaaS umbrella. So what do all of these acronyms stand for?
To quote Diamandis,
“ Infrastructure as a service (IaaS): … IaaS products allow clients to outsource physical or hardware infrastructure to the cloud. Think data storage and virtualization, remote servers, outsourced networking, and firewall services.”
“Platform as a service (PaaS): ….PaaS is a domain for software developers. Typically consisting of operating systems (OS) and developer tools, PaaS products help software engineers build, test, deploy, maintain, and manage every step of the development life cycle of web applications.”
“Software as a service (SaaS): …SaaS is a domain for end-users (or business clients). Offering on-demand, pay-per-use service of an application, SaaS products are accessible anywhere, and entirely platform-independent. As all computing resources responsible for delivering SaaS are managed by the vendor, users can thereby run the program on their own business data to replicate software-driven processes (whether in analytics, decision-making, optimization problems, etc.). As a result, brand new startups and large corporations alike are saved precious human capital and the costly, time-consuming development of in-house software.”
What about Artificial Intelligence as a service (AIaaS), Machine Learning as a service (MLaaS), and other acronyms? These different iterations of “X as a service” involve artificial intelligence and machine learning models made available to its users as a service. But how these models are offered and accessed adds another layer of confusion.
There are both SaaS Artificial Intelligence (AI) and Machine Learning (ML) offerings, meaning there is some kind of user interface or webapp that the user interacts with to upload data, access the AI and ML models, and generate new insights, as well as PaaS AI and ML offerings, meaning there is no webapp or user interface.
PaaS solutions may be accessed via an application programming interface (API) and are geared towards software developers. APIs allow the software developers to incorporate functionality, including AI and ML, into their apps, websites, etc. without the knowledge, skills, time, etc. to build their own functions. Why reinvent the wheel if you don’t have to? Widely known SaaS apps, such as Venmo, are often built on less widely known PaaS solutions such as the banking platform Plaid. These API platforms often run in the background without the user knowing.
“The advantages of PaaS are primarily that it allows for higher-level programming with dramatically reduced complexity; the overall development of the application can be more effective, as it has built-in/self up-and-down ramping infrastructure resources; and maintenance and enhancement of the application is thus easier.”
Foxy AI — A PaaS company?
Foxy AI, in its current form, is a platform as a service company. We build and provide computer vision models as a service to our customers. Our models are used to streamline workflows and reduce quality assurance overhead by automatically identifying data in property photos. These models can be simple, like determining if it’s an image of a kitchen or a bathroom, or more complex, like analyzing the quality and condition of the property. If we don’t have a model built to meet the immediate needs of our customers, we work with them to develop and deploy new models. So maybe this makes us a CV PaaS company?
Imagine if you had to gain a decade of AI and CV experience to add image search functionality to your website. It would be extremely cost and time prohibitive. By using Foxy AI’s services, software developers can leverage our collective expertise to quickly and easily incorporate artificial intelligence and computer vision into their applications, without the requisite experience.
Do we need more acronyms to specify if a company is an AI SaaS or an AI PaaS company? Maybe, but lets hope not!