free counter

Putting artificial intelligence and machine learning workloads in the cloud

Stephen Pritchard


Published: 19 Sep 2022

Artificial intelligence (AI) and machine learning (ML) are a few of the most hyped enterprise technologies and also have caught the imagination of boards, with the promise of efficiencies and lower costs, and the general public, with developments such as for example self-driving cars and autonomous quadcopter air taxis.

Needless to say, the truth is more prosaic, with firms seeking to AI to automate areas such as for example online product recommendations or spotting defects on production lines. Organisations are employing AI in vertical industries, such as for example financial services, retail and energy, where applications include fraud prevention and analysing business performance for loans, demand prediction for seasonal products and crunching through vast levels of data to optimise energy grids.

All of this falls lacking the thought of AI being an intelligent machine such as 2001: AN AREA Odysseys HAL. Nonetheless it continues to be a fast-growing market, driven by businesses attempting to drive more value from their data, and automate business intelligence and analytics to boost decision-making.

Industry analyst firm Gartner, for instance, predicts that the global market for AI software will reach US$62bn this season, with the fastest growth via knowledge management. Based on the firm, 48% of the CIOs it surveyed have previously deployed artificial intelligence and machine learning or intend to do so next 12 months.

A lot of this growth has been driven by developments in cloud computing, as firms may take benefit of the low initial costs and scalability of cloud infrastructure. Gartner, for instance, cites cloud computing as you of five factors driving AI and ML growth, since it allows firms to experiment and operationalise AI faster with lower complexity.

Furthermore, the large public cloud providers are developing their very own AI modules, including image recognition, document processing and edge applications to aid industrial and distribution processes.

A few of the fastest-growing applications for AI and ML remain e-commerce and advertising, as firms turn to analyse spending patterns and make recommendations, and use automation to focus on advertising. This takes benefit of the growing level of business data that already resides in the cloud, eliminating the expenses and complexity connected with moving data.

The cloud also lets organisations take advantage of advanced analytics and compute facilities, which are generally not cost-effective to create in-house. This consists of the usage of dedicated, graphics processing units (GPUs) and intensely large storage volumes permitted by cloud storage.

Such capabilities are beyond the reach of several organisations on-prem offerings, such as for example GPU processing. This demonstrates the significance of cloud capability in organisations digital strategies, says Lee Howells, head of AI at advisory firm PA Consulting.

Firms may also be accumulating expertise within their usage of AI through cloud-based services. One growth area is AIOps, where organisations use artificial intelligence to optimise their IT operations, especially in the cloud.

Another is MLOps, which Gartner says may be the operationalisation of multiple AI models, creating composite AI environments. This enables firms to develop more comprehensive and functional models from smaller blocks. These blocks could be hosted on on-premise systems, in-house, or in hybrid environments.

Cloud providers AI offerings

In the same way cloud providers offer the blocks of IT compute, storage and networking so that they are accumulating a variety of artificial intelligence and machine learning models. Also, they are offering AI- and ML-based services which firms, or third-party technology companies, can build to their applications.

These AI offerings need not be end-to-end processes, and frequently they’re not. Instead, they offer functionality that might be costly or complex for a company to supply itself. However they may also be functions which can be performed without compromising the firms security or regulatory requirements, or that involve large-scale migration of data.

Types of these AI modules include image processing and image recognition, document processing and analysis, and translation.

We operate in a ecosystem. We buy bricks from people and we build houses along with other things out of these bricks. Then we deliver those houses to individual customers, says Mika Vainio-Mattila, CEO at Digital Workforce, a robotic process automation (RPA) company. The firm uses cloud technologies to scale up its delivery of automation services to its customers, including its robot as something, that may run either on Microsoft Azure or perhaps a private cloud.

Vainio-Mattila says AI has already been an important section of business automation. One that is just about the most prevalent is intelligent document processing, that is basically making sense of unstructured documents, he says.

The target would be to make those documents meaningful to robots, or automated digital agents, that then do things with the info in those documents. This is the space where we’ve seen most usage of AI tools and technologies, and where we’ve applied AI ourselves most.

He sees an evergrowing push from the large public cloud companies to supply AI tools and models. Initially, that’s to third-party software suppliers or providers such as for example his company, but he expects the cloud solution providers (CSPs) to provide more AI technology right to user businesses too.

Its a fascinating space as the big cloud providers spearheaded by Google obviously, but very closely accompanied by Microsoft and Amazon, among others, IBM aswell have implemented services around ML- and AI-based services for deciphering unstructured information. Which includes recognising or classifying photographs or, or translation.

They are general-purpose technologies designed in order that others can reuse them. The business enterprise applications are generally very use-case specific and need experts to tailor them to a companys business needs. And the focus is more on back-office operations than applications such as for example driverless cars.

Cloud providers also offer domain-specific modules, in accordance with PA Consultings Howells. These have previously evolved in financial services, manufacturing and healthcare, he says.

Actually, the number of AI services offered in the cloud is wide, and growing. The big [cloud] players will have models that everyone may take and run, says Tim Bowes, associate director for data engineering at consultancy Dufrain. 2-3 years ago, it had been all third-party technology, however they are actually building proprietary tools.

Azure, for instance, offers Azure AI, with vision, speech, language and decision-making AI models that users can access via AI calls. Microsoft breaks its offerings into Applied AI Services, Cognitive Services, machine learning and AI infrastructure.

Google offers AI infrastructure, Vertex AI, an ML platform, data science services, media translation and speech to text, to mention several. Its Cloud Inference API lets firms use large datasets stored in Googles cloud. The firm, unsurprisingly, provides cloud GPUs.

Amazon Web Services (AWS) also offers a wide variety of AI-based services, including image recognition and video analysis, translation, conversational AI for chatbots, natural language processing, and a suite of services targeted at developers. AWS also promotes its health insurance and industrial modules.

The large enterprise software and software-as-a-service (SaaS) providers likewise have their very own AI offerings. Included in these are Salesforce (ML and predictive analytics), Oracle (ML tools including pre-trained models, computer vision and NLP) and IBM (Watson Studio and Watson Services). IBM has even developed a particular group of AI-based tools to greatly help organisations understand their environmental risks.

Specialist firms include, UIPath, Blue Prism and Snaplogic, even though latter three could possibly be better referred to as intelligent automation or RPA companies than pure-play AI providers.

It really is, however, an excellent line. In accordance with Jeremiah Stone, chief technology officer (CTO) at Snaplogic, enterprises tend to be embracing AI on an experimental basis, even where older technology could be appropriate.

Probably 60% or 70% of the efforts Ive seen are, at the very least initially, getting started exploring AI and ML in an effort to solve issues that could be better solved with an increase of well-understood approaches, he says. But that’s forgivable because, as people, we continually have extreme optimism for what software and technology can perform for all of us if we didnt, we wouldnt progress.

Experimentation with AI will, he says, bring longer-term benefits.

Cloud-based AIs limits and prospects

You can find other limitations to AI in the cloud. First of all, cloud-based services are suitable to generic data or generic processes. This enables organisations to overcome the security, privacy and regulatory hurdles involved with sharing data with third parties.

AI tools counter this by not moving data they stay static in the neighborhood business application or database. And security in the cloud is improving, to the stage where more companies are willing to utilize it.

Some organisations would rather keep their most sensitive data on-prem. However, with cloud providers offering industry-leading security capabilities, the reason behind achieving this is rapidly reducing, says PA Consultings Howells.

Nonetheless, some firms would rather build their very own AI models and do their very own training, regardless of the cost. If AI may be the product and driverless cars certainly are a prime example the business enterprise would want to own the intellectual property in the models.

But even then, organisations stand to reap the benefits of areas where they are able to use generic data and models. The elements is one of these, image recognition is potentially another.

Even firms with very specific demands for his or her AI systems might take advantage of the expansive data resources in the cloud for model training. Potentially, they could also desire to use cloud providers synthetic data, that allows model training minus the security and privacy concerns of data sharing.

And few in the market would bet against those services coming, first of all, from the cloud providers.

Read more on Artificial intelligence, automation and robotics

Read More

Related Articles

Leave a Reply

Your email address will not be published.

Back to top button

Adblock Detected

Please consider supporting us by disabling your ad blocker