Some of the largest and most innovative technology companies in the world are investing heavily in cloud artificial intelligence developer services, with the global AI software market expected to reach a whopping US$135 billion (NZ$214 billion) by 2025, according to IT analysts Gartner.
To build next-generation applications, developers need services that enhance app capabilities in the areas of automated machine learning, language and vision, Gartner said in its new 2022 Magic Quadrant for Cloud AI Developer Services.
The rise of AI and machine learning (ML) creates a challenge for software engineering leaders, as few developers are data science experts.
A 2021 Gartner survey found that more than 75 per cent of IT leaders had fewer than a quarter of their organisation’s software engineers trained in ML, showing the significant need for cloud-based developer service tools for software engineering teams.
Cloud AI developer services are cloud-hosted or containerized services that enable development teams and users who are not data scientists to use AI models via application programming interfaces, software development kits (SDKs) or applications.
These AI services help users provide services with capabilities in areas of automated machine learning (autoML), language and vision, such as natural language understanding (NLU), image recognition and machine leaning (ML) services.
In its Magic Quadrant report, the analysts identified and ranked the 13 leading providers worldwide of cloud artificial intelligence developer services paving the way for the future.
Gartner’s Magic Quadrant ranks vendors on their ability to execute and completeness of vision and places them in four categories:
- Niche players: low on vision and execution.
- Visionaries: good vision but low execution.
- Challengers: good execution but low vision).
- Leaders: excelling in both vision and execution.
The world’s largest software company won a gold medal for vision on Gartner’s Magic Quadrant and ranks second for execution.
Microsoft’s Azure AI platform has a comprehensive offering for language, vision and autoML use cases.
Its services can be consumed by professional developers via application programming interfaces and software development kits.
They can also be accessed via the Microsoft Power Platform.
The company leads the industry in terms of enterprise-quality properties, such as integration, scalability, performance, security, privacy, transparency, explainability and responsible use of AI.
Strengths: Microsoft differentiates its AI developer services by making it easy for both professional programmers and occasional developers to use machine learning, language and vision services in their applications.
Microsoft’s internal use of cloud AI developer services across its product lines provides a distinct advantage, enabling it to improve quality and time to market faster than its competitors.
Weaknesses: Microsoft offers its full portfolio of services in the Americas, but does not offer all these services elsewhere.
Gartner said potential customers should determine whether Microsoft supports these services in their region.
Leader: Amazon Web Services (AWS)
The world’s leading cloud computing provider won the gold medal for execution on Gartner’s Magic Quadrant and fourth for vision.
AWS provides AI services, including Amazon SageMaker and other popular language and vision services, designed to automate the full AI development and operationalisation cycle.
The cloud unit of online retail giant Amazon allows customers to build solutions on their own, with the assistance of dedicated AWS personnel, or with the help of consulting partners.
AWS is a top choice for production workloads, due to low operational costs and the breadth of its AI services and infrastructure.
Strengths: AWS excels in terms of AI operationalization and production scalability. Amazon SageMaker enablers developers to deploy trained models in production environments with a single click.
AWS is steadily expanding its presence in the cloud AI developer market by attracting hundreds of thousands of customers to its AI services.
Weaknesses: AWS is the only ‘Leader’ ranked in Gartner’s Magic Quadrant that is lacking a multi-cloud and hybrid-cloud vision.
Customers looking for multi-cloud and hybrid-cloud solutions from AWS have limited options and will likely need to partner with other vendors.
The search and cloud computing giant offers language, vision and autoML services via Vertex AI on the Google Cloud Platform (GCP).
Google’s services focus on deep neural network models.
The company also offers Contact Centre AI and Document AI, along with pretrained ML models that developers can customize.
It is a leader in AI research and responsible use of the technology; more than 3,500 researchers having published over 6,000 papers on AI related topics.
Google ranks second for vision and third for execution in Gartner’s 2022 Magic Quadrant for Cloud AI Developer Services.
Strengths: Google has well-defined ethics processes and can quickly address concerns about AI on a case-by-case basis.
It has model cards that explain the essential elements of an ML model, and fairness indicators that automatically assess bias in datasets and models.
Weaknesses: Google does not fully support deployment of AI services in private clouds or on-premises.
Google’s development efforts focus almost exclusively on neural networks, with little attention paid to symbolic AI.
IBM services span all segments of the cloud AI developer services market, ranking third for vision and fourth in execution on Gartner’s Magic Quadrant.
The company has consolidated its AI offering under the Watson brand, and refined its requirements for products and services to clarify its positioning.
IBM has strengthened the integration between its industry-leading research division and its product organisations to ensure their innovations are added to products in a timely fashion.
It leads with a true hybrid cloud strategy that appeals to most customers.
Strengths: IBM’s language services are especially strong.
Developers can integrate IBM Watson Assistant, Watson Natural Language Understanding and Watson Discovery to streamline discovery and creation of intents and entities.
Weaknesses: Although IBM’s offerings are comprehensive, they lack in some areas, such as image labelling and generation.
Additionally, some customers have stated that IBM’s pricing is excessive, with high transaction costs.
Challenger: Alibaba Cloud
As one of the largest cloud players throughout Asia-Pacific, China's Alibaba Cloud offers a complete set of cloud AI developer services spanning language, vision and autoML services.
The cloud provider packages its services in a number of ways to suit professional and independent developers.
The company ranks fifth for execution and sits in the middle of the pack in Gartner’s Magic Quadrant.
Strengths: An AI marketplace that offers more than 1,600 models.
Alibaba Cloud offers flexible pricing: either pay-as-you-go or periodic charges.
It also provides some free APIs for developers.
Weaknesses: About 99 per cent of its cloud AI developer services clients are based in China.
Although Alibaba Cloud provides comprehensive customer support, it has fewer cloud AI developer services partners than its major competitors.
China-based Baidu offers an extensive range of AI services for autoML, language and vision functions.
Its Baidu Brain services support its internal AI and its commercial AI services.
Baidu has more than 1,300 employees who focus on commercial cloud AI developer services.
At its AI Technology Group and AI Labs, Baidu has about 2,000 employees who develop technologies such as open-source deep learning frameworks, AI chips, virtual assistants and autonomous driving.
The company ranks in the middle of the pack for both vision and execution on Gartner’s Magic Quadrant.
Strengths: Gartner said Baidu is a leading AI innovator.
As of November 2021, it had more AI patents in China than any other company around deep learning, language, vision, autonomous vehicles and industry-specific solutions.
Weakness: Baidu’s AI solutions are geared primarily to developers in China.
Some of its natural language offerings, like text analytics and sentiment analysis, are available only in Chinese.
Known for being the world’s largest gaming company and for WeChat, China's Tencent offers cloud AI services for vision, language and autoML use cases.
Tencent launched its AI Lab in 2016, followed by opening an AI research center in Seattle.
Its innovation, implemented solutions, consumer platforms and array of multimedia data enable Tencent to develop advanced scalable capabilities.
Tencent has more than 1,170 employees working on cloud AI developer services.
Gartner ranks the company among the middle of the pack for both vision and execution.
Strengths: Tencent’s YouTu Lab is one of the leading AI research centers focused on vision services and deep learning.
Its image recognition function allows users to identify the specific brand, name, model or style of a product in an image and to generate the product’s price.
Weaknesses: Regarding its language and autoML services, Tencent focuses solely on the Chinese market, and has limited support for other languages.
For example, its sentiment and text analytics services are available only in Chinese.
The US enterprise software giant is broadening its portfolio of AI services beyond Oracle Digital Assistant and its other applications to provide autoML, language and vision services.
Oracle has consolidated its AI portfolio under a newly formed team and has an aggressive plan to expand its offerings.
Its strengths reside in its language-focused conversational AI platform and its NLU and text analytics.
Oracle ranks fifth for vision on Gartner’s Magic Quadrant and near the bottom for execution.
Strengths: Oracle has a strong developer outreach program and following among the developer community.
The company has strong NLU capabilities, with its NLU product being among its most mature offerings.
Weaknesses: Although Oracle offers competitive AI services, there are many gaps in its offerings.
To compete in the cloud AI developer market, Gartner said Oracle must quickly expand beyond its limited range of AI services.
Niche Player: Dataiku
US company Dataiku’s platform provides strong autoML services and some language and vision services, and offers data, analytics and AI designed for a variety of users.
Dataiku is a new entrant to Gartner’s Magic Quadrant for Cloud AI Developer Services.
Developers to build data pipelines, data visualisations, ML and deep learning projects, and end-user applications with Dataiku technology.
It emphasises its platform’s ease of use, time to value, composability and support for collaboration between various roles in order to deliver AI solutions.
The company ranks in the middle of the pack for both vision and execution on the Magic Quadrant.
Strengths: Dataiku customers have praised its customer support.
The company offers concrete service level agreements for all customers at no additional cost based on classification of errors.
Weaknesses: Known for its data science and ML platform, Dataiku has yet to gain equal recognition from AI developers, despite developer-friendly features.
There are also functionality gaps in Dataiku’s language and vision services.
Niche Player: H2O.ai
H2O.ai offers a platform that provides language, vision and autoML services that can run in cloud, on-premises, edge and hybrid environments.
The H2O AI Cloud grew tremendously in 2021, Gartner said, as the company remains a thought leader in autoML across structured, time-series, image, video, audio, text and document data.
H2O.ai is a major open-source contributor with 20,000 companies supporting H2O.ai’s open-source offering and more than 1 million users.
It ranks among the middle of the pack for both vision and execution on Gartner’s Magic Quadrant.
Strengths: H2O.ai has been aggressively revamping its sales approach.
It has introduced a consumption pricing model, whereby customers consume cloud AI units that can be applied to any H2O AI Cloud service.
Weaknesses: H2O.ai has resource constraints that affect its ability to cover global operations, especially in regions where its presence is growing.
Niche Player: Clarifai
Clarifai is known for image services and has been a strong provider in that area for many years.
The company’s cloud AI developer services offering and roadmap span language, vision and autoML use cases.
Although Clarifai grew significantly in terms of sales and partnerships in 2021, it is still a midstage startup and not yet profitable, Gartner said.
The company ranks near the bottom of the pack for both vision and execution on the Magic Quadrant.
Strength: Clarifai invests more than half its revenue in research and development.
Planned innovations in language services include zero-shot text classification, knowledge graph linking and speaker identification.
The company also plans to develop distributed learning and federated learning capabilities for the autoML use case.
Weakness: Clarifai is a relatively small, venture-funded company that is competing with the cloud giants.
Clarifai does not have personnel in all the regions where its customers are located.
Niche Player: Prevision.io
The French company’s platform focuses on helping data scientists and software engineers develop, operate, maintain and manage ML models using autoML services.
Prevision.io is investing in improving its text to speech, speech to text, translation and generation capabilities within its language services portfolio.
For its vision service offering, it is enhancing its video content analysis and labeling capabilities.
Prevision.io ranks land in vision on Gartner’s Magic Quadrant and near the bottom for execution.
Strengths: Prevision.io’s revenue grew by 60 per cent year over year in 2021, with a strong sales pipeline.
The number of active users on the Prevision.io platform also increased significantly.
Weaknesses: Prevision.io has a full-featured autoML offering, but its language and vision services are among the least capable offered by vendors in Gartner’s list.
Customers will need to depend on other vendors for language and vision capabilities to complement Prevision.io’s autoML platform.
Niche Player: Aible
Aible provides ROI-centered autoML services, as well as language and vision services.
The company’s platform has ML, optimization, simulation, recommendation and monitoring capabilities.
All offerings are serverless-first and models can be deployed on Microsoft Azure and AWS.
The platform is easy to use, guided and engaging, while allowing developers to use no-code, low-code and code-based approaches.
Aible ranks last in execution on Gartner’s Magic Quadrant and near the bottom for vision.
Strengths: To create AI strategies that deliver business impact, Aible starts by understanding its customers’ business goals and engaging all stakeholders.
The company delivers an intuitive, end-to-end experience that is driven by data, what-if testing and scenario planning.
Weaknesses: Despite Aible’s growth and recognition, it is less well-funded and recognized than its larger and more established competitors.
Aible’s geographic presence is limited to four offices in North America and two offices in Europe.