Why Generative AI Company Ecosystem is More Than Just Chatbots and Prompts
If you’ve been keeping up with social media recently, you may have noticed some of your friends are posting photos of themselves as their own AI avatars or talking about how they used chatbots to help them write a difficult email to their boss. It’s almost like magic, right? But when you start looking at the details, it becomes clear that this “magic” is coming from multiple sources. In technology, we refer to this as the generative AI company ecosystem. An analogy is the automotive industry and the repair process. When you take a ride in your car (the app), there are typically many people involved in making your car run smoothly. As well as many businesses supplying the raw materials (the supplies), and numerous gas stations that enable you (the fuel) to operate. Currently the generative AI company ecosystem is growing at such a rapid pace that even those working within those industries cannot keep up with the pace of everything that is happening within the industry.
The Foundation: Why chips and clouds are the new oil
Breaking down the AI infrastructure and platforms that keep the lights on for every modern AI tool.
The Brains of the Operation: Foundation models explained
How massive “General Purpose” models act as the core intelligence for the entire generative AI company ecosystem.
The Toolbelt: AI creative tools and niche applications
Why the application layer is where everyday users finally get to play with the power of AI.
Survival of the Smartest: AI industry competition trends
A look at where the money is flowing and which parts of the value chain are actually sustainable.
Understanding AI infrastructure and platforms heavy lifting
Land and utilities are prerequisites to constructing a home. Similarly, in the generative AI ecosystem, chips (specifically GPUs) and huge data centers constitute the land. I’m sure you heard the name ‘Nvidia’, as the stock price reflects, it tells all about the entire AI infrastructure. They provide the raw computing resources. Not solely do you need the hardware, but you need cloud providers like Microsoft Azure, Google Cloud and Amazon Web Services (AWS). Which supply you with platforms where those AI models can live. Without these behemoths, you don’t have anywhere to run your favorite AI. Consider their platforms to be rented laboratory space to use for your experiments. For companies in Asia, particularly scalable companies, choosing which of these cloud providers. Those utilize as the basis for their building will be among the greatest and costliest business decisions ever made. This layer represents the foundation on which the AI’s value chain analysis will occur. And it is the source of the majority of “guaranteed” income in the AI industry today.
Foundation models in AI
After you’ve gained strength, it’s time to develop intelligence. Foundation models in artificial intelligence serve this purpose. Think of a foundation model like a student who has read every single book in the library. They have a general knowledge of many things, but they aren’t particularly skilled at doing anything yet.
The large, intelligent foundation models are being developed by companies such as OpenAI (with GPT-4), Google (with Gemini) and Anthropic (with Claude). These models form the foundation of the generative AI company ecosystem. As nearly all other companies build their technologies on top of these foundation models. The cost to develop these foundation models is astronomical. And also require hundreds of thousands of specialized chips running for months to accomplish. This is why we don’t see a new “GPT” level model produced on a regular basis. The barrier to entry is mind-bogglingly high. It’s the “Big Tech” playground, and the AI industry is currently filled with just a handful of “Big Tech” corporations. That are financially capable of paying the electrical bill associated with operating these foundation models.
Generative AI company ecosystem application layer platforms
This is what we see. Most people don’t care about chips or raw model parameters. They just want to know whether AI can assist them with designing a logo or summarizing 50-page PDF document’s summary. This is the AI application layer platforms & AI Creative Tools ecosystem. Like Canva’s AI features, & tools such as Jasper & Copy.ai rarely have large models built from the ground up. They typically “rent” foundations model providers’ “brains’ , add a “user friendly” skin on the outside to provide an-to-use product for the specific task(s ) required. As example, ensuring the AI learns & understands the subtle “nuances” associated with marketing slang in Malaysia . Other than that, specific formatting requirement when preparing an accounting report. This is where we see the majority of creative explosion from generative AI companies that bridges the gap between high level technology & your day-to-day life in KL or Singapore.
Generative AI market trends and competition
What’s going on now? Trends indicate that generative AI is being split into general-use AIs and AIs used for specific applications. Which with people moving from being in awe of computers being able to converse with them to having a need for them to be dependable. Also, the AI industry competition is moving away from the “size of the model” competition and into an environment where the “best data” wins as evidenced by an attorney. They using a specialized AI model that has been specifically trained on documents pertaining to the laws of Malaysia. As they opposed to using a general AI model that has been developed in Silicon Valley.
There is also a trend toward creating a diverse set of AI platforms. As seen in the rise of “Vertical AIs,” or AIs built to operate primarily in specific industries such as healthcare or logistics. An observation that can be made with regards to these trends in Asia is how they may adapt to these global technologies. Many smaller start-ups are succeeding by leveraging the “application layer” of technology. While competing with the “model wars” being played out in the USA and China by developing tools that make existing, large models. Such as GPT, usable for a specific business function.
Why the generative AI company ecosystem matters to you
Knowing about the generative artificial intelligence (AI) technology company ecosystem will help show any hidden “hoopla” that comes out when a new generative application releases. When you see a new application released, you can ask, “Are they creating something new?” or “Are they simply using a beautiful user interface on top of an existing Generative AI model?”
This is a very important point, not just “some geeky tech”. The “engine manufacturers” such as Nvidia and Open AI are the major drivers of the new generative AI industry. They will provide power to the “vehicle designers”, or application layer, that actually determine and shape how we interact with this new world of technology. Due to the rapid evolution of the industry and a multitude of competing solutions, there will continue to be a great many products and services that will provide limitless opportunity. It does not matter if you are a business owner in Penang or an application designer in Singapore, the most important point is to know what your role in this ecosystem is.
