Blog post

Gartner Launches Artificial Intelligence Emerging Technologies Radar

By Anthony J. Bradley | October 27, 2020 | 3 Comments

Tech and Service Providersemerging technologiesArtificial IntelligenceEmerging Technologies and Trends Impact on Products and Services

Last week we published a new Gartner Emerging Technologies and Trends Impact Radar for Artificial Intelligence (full report available to subscribing Gartner clients). In this research we highlight 24 AI-related emerging technologies and four overarching trends. 

The four overarching trends include:

  • Artificial intelligence (AI) developer toolkits, services, marketplaces and easy-to-use APIs are beginning to “democratize” AI.
  • Growing AI adoption is beginning to shift business automation from process automation to intelligent business automation.
  • Advanced hardware, innovative software techniques and micro-AI are accelerating Edge AI adoption moving more and more processing from the cloud to the edge.
  • AI (advanced NLP) is actually starting to transform how humans and machines interact. 

In the Artificial intelligence Emerging Technologies and Trends Impact Radar, the rings represent the Range. Range estimates the distance (in years) that the technology or trend is from “crossing the chasm” from early-adopter to early majority adoption. The size and color of the emerging technology or trend radar blip represents its Mass. This indicates how substantial an impact the technology or trend will have on existing products and markets.

Here is our 2020 artificial intelligence emerging technologies impact radar (click to enlarge).  

Let’s look at a few that I find especially interesting. I’ll examine a few that are right around the corner (AI developer and teaching kits, transformer-based language models, intelligent applications) and one that is further out (AI-Generated Composable Applications).

AI Developer and Teaching Kits

Artificial intelligence (AI) developer and teaching kits are instructions, examples, tools and software development kits (SDKs). They provide an abstraction layer on top of data science platforms, frameworks, analytic libraries and devices. And so they make it faster and easier for software engineers to build AI into applications. 

These kits are 1 to 3 Years from early majority because they address a subset of needs including kits for virtual assistants, AI design kits, and AI-mobile-serving SDKs. Also, dependencies associated with vendor specific toolkit offerings limit usability.  Additionally, there are challenges with scaling outputs and the cost of integrating different formats and interfaces across the stack. 

Their impact mass is high because AI developer and teaching kits are a significant accelerator to AI adoption. This enables a much larger set of software developers to more effectively and efficiently contribute to AI development and implementation. Over the next three years, AI developer and teaching kits will provide a strong foundation for the expansion of more  complex AI-enabled capabilities.

The full analysis in the document was authored by Eric Hunter.

Transformer-Based Language Models

Transformer-based language models are DNNs that process words as sequences in a sentence. This approach preserves the context or meaning surrounding terms. It also substantially improves translation, transcription and natural language generation. These models are trained on enormous data sets of billions of phrases. Example transformer-based models include BERT, BART and GPT-3.   

The 1-3 year range is driven by the effectiveness of the training tools, the runtime efficiency and the ease of deployment. Transformer-based language models such as GPT-3 have the capability to generate paragraphs of text that are indistinguishable from those written by a well-educated human. 

The impact mass of transformer-based language models is very high because they are displacing RNN systems at a surprising rate. And new tools deliver substantial improvements in advanced text analytics and all the related applications such as conversational user interfaces, intelligent virtual assistants and automated text generation. 

The full analysis in the document was authored by Martin Reynolds.

Intelligent Applications

Intelligent applications are enterprise business applications with embedded or integrated artificial intelligence technologies, such as intelligent automation, data-driven insights and guided recommendations. They represent a transformational shift in business applications from primarily procedural tools that help execute tasks to intelligent software that also assists in acquiring knowledge, visualizing key data and advising on relevant decisions.  

Intelligent applications are 1-3 years from crossing the chasm because many of the large software vendors are now embedding AI into their products. And their efforts will create competitive momentum further driving adoption across application domains such as enterprise resource planning (ERP), sales force automation (SFA), HR and customer relationship management (CRM). Intelligent applications are the next major battleground for enterprise application providers and it will be many years before we hit the top of this s-curve. 

The impact on existing technologies is high because it refactors enterprise applications. This is a competitive opportunity for new entrants into the market. It also creates potential for existing players to gain, or lose, market share as competitive advantage shifts to intelligent application capabilities. 

The full analysis in the document was authored by Alys Woodward.

AI-Generated Composable Applications

AI-generated composite applications build business applications using artificial intelligence to assemble application components (without human developer involvement) to meet new and even ad hoc business needs. Context-aware AI will detect a specific business need in response to a business situation and automatically assemble the application using packaged business capabilities (PBCs) as building blocks. 

This technological capability is 6 to 8 Years from crossing the chasm because it is dependent on the emerging trend where technology providers shift from delivering large and mostly static business applications to offering smaller PBCs with robust application programming interfaces (API). Most of the technology already exists (e.g., APIs, microservices, self-integration, containerized software). But pulling it all together into a composable applications ecosystem with standards that facilitate interoperability is still a long way off. In this instance AI technology advancement is not the primary inhibitor.        

AI-generated composable applications will have a high impact on the entire application software market and businesses across industries and geographies. They represent a substantial improvement in business agility allowing businesses to respond more quickly to changing technology and business situations. The entire business world will be able to move faster which can lead to even more rapid change across business and society. Currently, Gartner is not aware of any vendor offering AI-generated composite applications in the market. 

The full analysis in the document was authored by  Jim Hare and Paul Saunders.

A New Series of Research

The Artificial Intelligence Emerging Technologies Impact Radar is one impact radar in a series of 15 impact radars including these (available to subscribing Gartner Clients):

Always interested in your feedback on our research so we can continuously improve what and how we deliver.

And don’t forget, if you are a technology product or service leader and a Gartner client, don’t miss out! Subscribe to Gartner’s Emerging Technologies and Trends Research. These Impact Radars are another example of the new content we are producing.


A big THANK YOU to the well over 100 Gartner analysts and associates who worked on this series! All the organizing, managing, brainstorming, writing, reviewing, editing, graphics, etc. was a tremendous multi-month effort. Big shout out to Errol Rasit for managing delivery of the whole series.  

A special thanks, for this Artificial intelligence Emerging Technologies Impact Radar, to analysts Annette Jump and Danielle Casey for leading the effort and to co-author analysts Eric Goodness, Alys Woodward, Erick Brethenoux, Martin Reynolds and Jim Hare. I personally appreciate your patience with me through what probably felt like a stream of never ending edits, suggestions and comments.

The Gartner Blog Network provides an opportunity for Gartner analysts to test ideas and move research forward. Because the content posted by Gartner analysts on this site does not undergo our standard editorial review, all comments or opinions expressed hereunder are those of the individual contributors and do not represent the views of Gartner, Inc. or its management.

Leave a Comment


  • John Saylor says:

    Great article.

    AI-generated composite applications.
    Like “Context-aware AI will detect a specific business need in response to a business situation and automatically assemble the application using packaged business capabilities (PBCs) as building blocks.”

    Like “Advanced hardware, innovative software techniques and micro-AI are accelerating Edge AI adoption moving more and more processing from the cloud to the edge.”

    Additional considerations.
    Predictive Security, with Edge AI is extremely important as the more distributed Edge Computing becomes the more important is the knowledge of predictive threat intelligence.

    Sustainable Availability, with Edge AI is also important in remote or off-grid implementations. Finding power grid sources from renewable energy in the natural environment setting makes Edge AI environmentally responsible.

    In disruptive times needs do not disappear – they shift. To thrive with disruption, you must shift yourself and your organization.

  • Mohit Shah says:


    With the mention of AI-generated composite applications for enterprise benefit, what is the testing approach for such apps for better quality and performance assurance?

    Also, can you elaborate- “Context-aware AI will detect a specific business need in response to a business situation and automatically assemble the application using packaged business capabilities (PBCs) as building blocks? “.

  • Anthony J. Bradley says:

    Thanks Mohit. Testing needs to be based on use cases vs. systems as executing on the use case can cross multiple systems. Context-aware AI will understand the broader implications of the intent and build an appropriate business process “on-the-fly.” It will then “orchestrate” pulling together the right systems resources to execute on the ad hoc process.