The First Truly Global Generative AI Landscape 2024
Check out AIport's first research effort into global GenAI developments
Generative AI (GenAI) has clearly been a global hot topic across industries for the past year or so. The release of ChatGPT, specifically, ignited immense public interest and excitement in GenAI and awakened us to its transformative potential.
While we believe ChatGPT is only a glimmer of what GenAI could possibly do, the world has witnessed an overwhelming GenAI penetration in downstream tasks since its release in Dec 2022 — business intelligence, education, synthetic data generation, personalization, design, customer satisfaction, creativity, and more.
To showcase the wide development of GenAI technology lately, a number of VCs and leading analytical agencies (such as Sequoia Capital, Antler, and Base10) have released GenAI landscapes, which highlight key players shaping this field.
While we truly appreciate these steps taken to highlight rapid GenAI development, we noticed that most reports gravitate toward the tech giants of Silicon Valley and the tech powerhouses of Europe. In other words, none of them are truly global, considering that the Global AI Index, as of 19th Feb 2024, includes 62 countries.
Of course, it is not necessary that all “AI countries” must be actively building GenAI solutions. Yet, in our research, we noticed that GenAI is rapidly developing in many countries outside Europe and North America. But these developments are often neglected in such landscapes, which typically cover up to 10 countries.
However, in our research, we found 35 countries who have made any significant stride in GenAI ever.
Thus, aligning with our goals at AIport, we intend to fill this gap and provide a truly global view of modern GenAI developments.
Here’s Volume 1 of our GenAI landscape:
To explore the image in more detail, follow the link.
To see the full list of models and their types, download the file below.
Our observations
The prevalence of North American companies in the landscape underscores its position as one of the foremost players in GenAI development and deployment globally. We found a total of 38 companies from North America, which is, by far, the most in comparison to any other continent in the world.
In Europe, a vibrant landscape of AI innovation is evident across most of its countries, with the UK, France, Germany, Netherlands, Austria, and Russia being the top 6 contributors of this region from the 17 European countries we considered in our research.
Unlike Europe and North America, we found that Asia boasts only a few globally renowned players (at least at the moment), with Israel, China, India, Singapore, and Japan being the top five of the 11 countries we found to have made any contributions to GenAI.
Despite that, we did not find this to have any effect on the continent’s global contribution in terms of the total number of unique companies developing their own GenAI solutions, which is 28 (compared to 38 in North America and 27 in Europe).
In other parts of the world we looked at, namely Africa, Oceania, and South America, we found only five countries represented by eight companies that have developed GenAI solutions. These are Kenya, South Africa, New Zealand, Australia, and Argentina. But it is somewhat justified considering that very few countries from these continents are included in the GlobalAI index in the first place.
Overall, a total of 11 companies worldwide have developed more than one type of GenAI model. Stability AI leads with five distinct GenAI model types (image, video, audio, 3D, and code), followed closely by OpenAI (chatbot, audio, video, and multimodal) and Google (text, image, audio, and multimodal) — both with four model types. Microsoft, Meta, Tencent, Baidu, and Yandex are among those companies that developed between two to three types of distinct GenAI models.
13 companies in our study have developed multiple models within a single GenAI category. AssemblyAI has two speech-to-text models, MosaicML offers two iterations of its MPT for code generation, while IPOXCap has introduced two chatbots designed for business intelligence applications.
Approximately 10% of all companies covered in the study have implemented multimodality in their GenAI models (systems that can process various types of data, such as text, images, audio, and video, simultaneously). The majority of these developers are located in the US: we found eight multimodal models from North America but only two from Europe, yet. This indicates that while multimodality represents an emerging trend, its adoption outside North America still remains in the nascent stages.
We also found Europe to have significantly lower contributions towards Code GenAI in comparison to both North America and Asia. In fact, North America is the only region that covers at least one model from each of the 10 model categories depicted in the landscape. However, we did discover categories where Europe stands out, business intelligence and text being two of them.
Similar to North America, Europe possesses an almost equally diverse coverage across most GenAI categories (all categories except gaming, to be specific) in our landscape.
Our research and categorization methodology
To respect the “globality” baseline, we took all 62 countries that are included in the Global AI index.
Next, we researched prominent players who have developed in-house GenAI solutions in these countries. Here, we only found 35 such countries, which is still pretty good, given the fact that GenAI only actively kicked off in late 2022.
We put these players, based on what kind of GenAI solutions they are developing, into 10 categories — text, audio, video, chatbot, image, gaming, business intelligence, code, 3D rendering, or multimodal.
The categorization was based on the type of output produced by the model. For instance, speech-to-text models were categorized as “Text”, text-to-image models were categorized as “Image”, and so on.
Here’s how many companies we considered from each region:
38 from North America
28 from Asia
27 from Europe
4 from Oceania
3 from Africa
1 from South America
Naturally, we realize that the GenAI market is developing very rapidly and there are many more companies out there, but we tried to include as many of the most prominent players that develop their own GenAI as possible.
This is why we’re calling this edition Volume 1 of the First Global GenAI landscape. We see it as an ongoing effort and aim to improve it and cover more companies over time.
If you’re a GenAI company and would like to be featured in the upcoming volumes of our landscape, or want to recommend a company you think should be featured, reach out to us at observanteditor@aiport.tech.
Final words
We at AIport firmly believe that of all the technology waves we have witnessed in the past — personal computers, dot-com, cloud computing — GenAI is the biggest of them all when it comes to transformative value and what it could do to the world.
In our opinion, what makes it so special and transformative in comparison to all other past waves is that this specific technology is about replicating human capabilities.
In other words, the GenAI technology, in general, resonates and does things that closely mimic how humans perform those activities, but less laboriously:
Code reviews, which were earlier done by reviewers manually, can now be done by GenAI.
Realistic image generation in seconds by simply providing a prompt, which, previously, took an artist many hours to produce.
Customer grievance resolution, which used to take hours due to manual involvement, is now being resolved in minutes.
The vast potential of GenAI has also been recognized by business. According to a recent report by Accenture, 97% of executives and business leaders said that they understand the potential of GenAI and what constructive value it could possibly add to their businesses. More specifically, leaders believe that there’s an immense value proposition that they can leverage from generative AI, specifically due to its automation and assistive capabilities.
Of course, there’s a lot to explore yet; the technology is pretty much in the early stages, and there’s a lot for us to figure out in terms of how we shall efficiently extract that big value, at scale.
We are eager to see how this space unfolds in 2024.
I've written about the Landscape report in my weekly newsletter. Here's the link: https://aiforreal.substack.com/p/when-gen-ai-breathes-life-into-robots
Excellent! Thank you.