By Vishal Goyal
It was in 1956 when the term “Artificial Intelligence” was coined, which means it has been existing for more than 60 years now. Still the academia and industry has started knowing its use and power in last few years only. We now see universities in India and across the globe including specialized courses and full time degrees in Artificial Intelligence at graduate level. Industries have been hiring data engineers and upskilling employees on areas of machine learning, deep learning, analytics etc. Below are few reasons for this buzz
Computational Power – With availability of high-end processing infrastructure (GPUs, TPUs etc), this has become much easier to run and train complex models much faster. Intel continues to lead chip designing with increased power and reduced size. Cost of owning and using such infrastructure continues to go down. Cloud based “Pay-As-You-Go” subscription models makes it available to even small and medium enterprises to explore and build solutions.
Data Availability – There was never a time when so much of data was available to us. For AI, data is food, which it needs to develop its muscles and show its strength. It is becoming impossible for organizations to make sense out of this data
Investments – Large corporates and venture capitalists continue to throw in money to see how AI can help solve complex business problems. IBM, Microsoft, Google, Amazon, NVDIA and other cash rich companies continue to invest in AI
AI as a Service (AaaS) – Microsoft (Azure), Amazon (AWS) and Google (GCP) through there platforms have built hundreds of cognitive services which has democratized the usage of AI for everyone to get started on their own. Create an account and get started. It is as simple as that now.
Year 2019 has already seen great use of AI in different fields and solving real business problems across industries like Healthcare (Cancer, Heart Disease detection), Manufacturing (Optimization solutions), Retail (Improving User experience) and Agriculture (Improving crop yield).
2019 and early 2020 will continue to see focus in below areas
Chatbots, Virtual Agents or Digital Assistants, terms synonymously used, is one of the hot topic in the market right now.
Availability of strong NLP (Natural Language Processing) engines, which are both open source and commercial, speech recognition libraries, high-end mobile phones and success of Alexa and Google Assistant is making implementation of chatbots far easier and quicker.
Technology giants such as Amazon, Google and Microsoft have been investing a lot into their AI engines and offering specialized services for developing chatbots through their cloud platforms. It is likely in the medium-to longer term these models will win out over the smaller vendors and the knowledge, dialog flows and intent models will leverage these platforms instead.
Commercial AI and ML
While most organizations still leverage open source technology to get started and building proof of concepts in AI, industries are looking at cloud based commercial solutions available for production readiness and deployments. These vendors provide fully managed service offerings from a fully scalable, secured platform without requiring complex skills to develop and understand the different components involved in building an AI solution. Amazon SageMaker, IBM Watson, SAS Viya, Azure ML Services are already offering production ready platform. SAP and Oracle have also significantly revamped their current ML strategies and offerings and moved away from being known as just “ERP” vendors to be seen as “Intelligent Enterprises”.
2019 will see increased usage of commercial platforms, which have integrated open source technology into their platforms and made is simpler for industries to consume these services
Given the amount of data, which exists today, exploring all possibilities becomes impossible. Data scientists are looking at use of automated algorithms to explore more hypotheses. Data science and machine learning platforms have transformed how businesses generate analytics insight.
Term “Citizen Data Scientists” is being used more commonly now. The idea is that organizations can leverage internal skills to bring the basic data science expertise in-house for advanced analytics while minimizing the burden on organizational resources. They can use these skills to wring more value from the analytics solutions they are using, as well as dig more deeply and broadly into data made available to them.
Augmented analytics identifies hidden patterns while removing the personal bias. Citizen data scientists use AI powered augmented analytics tools that automate the data science function automatically identifying data sets, developing hypothesis and identifying patterns in the data. Businesses will look to citizen data scientists as a way to enable and scale data science capabilities. Between citizen data scientists and augmented analytics, data insights will be more broadly available across the business, including analysts, decision makers and operational workers.
AIOps – Artificial Intelligence for IT Operations
Digital business transformation and speed at which needs to be done is necessitating a change to traditional IT management techniques. Consequently, we are seeing a significant change in current IT Ops procedures and a restructuring in how we manage our IT ecosystems.
Implementation of AIOps requires move away from siloed IT data in order to aggregate observational data (such as that found in monitoring systems and job logs) alongside engagement data (usually found in ticket, incident, and event recording) inside a Big Data platform (examples include Hadoop 2.0, Elastic Stack, and some Apache technologies). AIOps then implements Analytics and Machine Learning (ML) against the combined IT data, which provides continuous insights that, can yield continuous improvements with the implementation of automation.
AIOps bridges three different IT disciplines—service management, performance management, and automation.
AIOps will become more and more important and critical for IT operations to ensure 100% uptime, 24*7 operations and reduced costs of operations.
Biases point to the underlying ethics of AI. Algorithms learn from data and data is sometimes based on human decisions that are not always fair. Being able to show that automatic decisions taken by AI systems are based on underlying fair appraisals is crucial from this point on. This is being called “Explainable AI”.
So much so, that we expect to see a rising demand for regulation of AI. Transparency and ethics are one aspect of this debate, but AI in GDPR (which is already applicable) and other such regulations hold the potential to disrupt the widespread development of AI.
Edge Computing – IoT and AI
In 2019, AI models trained in the public cloud will be deployed at the edge. Industrial IoT is the top use case for artificial intelligence that can perform outlier detection, root cause analysis and predictive maintenance of the equipment.
Advanced ML models based on deep neural networks will be optimized to run at the edge. They will be capable of dealing with video frames, speech synthesis, time-series data and unstructured data generated by devices such as cameras, microphones, and other sensors.
Internet of Things is becoming the biggest driver of artificial intelligence in the enterprise. Ray Kurzweil, Google’s chief engineer, said AI would enhance, rather than displace, humanity. We must continue to embrace AI and see this as “Augmented Human Intelligence”.
(The author is the Head of Digital COE under LOS business unit Fujitsu Consulting India)
If you have an interesting article / experience / case study to share, please get in touch with us at [email protected]