It is capturing the world’s attention and involves many stakeholders including research, academia, industry, practitioners, policy makers and ethics advocates. AI is expected to be one of the most crucial enabling technologies in our lifetime.
AI is transforming industries, through the evolution of IT usage. Initially viewed as a tool to increase efficiency within organizations, for example, the use of computers to develop memos, the next inflection point saw IT becoming essential to measuring an organization’s performance against key performance indicators (KPIs) established by the management team.
The advent of the industrial internet of things (IIoT) saw IT go deeper into the management chain, all the way to decision makers and across more traditional industries that may not have been as reliant on IT in the past.
Today, AI is transforming the role of IT from one of measurement for the management team (performance relative to established KPIs), to one of providing insights to establish future goals and KPIs. Put simply, AI is taking a seat at the management table, adding its voice to where the organization should go via insights.
AI is already used in many applications, including healthcare for customizing patient treatments, the financial sector for fraud detection, autonomous vehicles for determining optimal speed, following and breaking distances and collaborative robots, designed to work safely alongside humans, lifting heavy loads, staging materials for human assembly, or completing repetitive motions.
Against this backdrop, in 2017, IEC and ISO became the first international standards development organizations (SDOs) to set up a joint committee (ISO/IEC JTC 1/SC 42) which will carry out standardization activities for artificial intelligence.
e-tech caught up with Wael William Diab, Chair of SC 42, following its inaugural meeting in Beijing this April. Diab is a business and technology strategist with more than 875 patents to his name in the field of information and communication technologies (ICT). He is currently a Senior Director at Huawei.
“One of the unique things about what IEC and ISO are doing through SC 42 is that we are looking at the entire ecosystem and not just one technical aspect. Combined with the breadth of application areas covered in IEC and ISO technical committees (TCs), this will provide a comprehensive approach to AI standardization with IT and domain experts.”
Diab explained the importance of taking a horizontal systems approach by working with as many people as possible, across IEC and ISO TCs, citing some examples of other JTC 1 SCs – internet of things, IT security and IT governance – the IEC Systems Committee for Smart Cities, and with external organizations. The key will be to get better leverage of liaisons and how to coordinate the work, so as to build on what already exists rather than duplicating it.
“This list will grow because the application domains are quite expansive, from digital assistants in smartphones to less obvious areas like online shopping market intelligence for determining a new market for a product, or healthcare, or the example of deciding whether someone will get a loan. All of these examples use learning algorithms.”
Another key area Diab highlighted was manufacturing and robots that help in the plant. Robots and humans that work side by side on an automanufacturing line and all the way through to deep analytics, means having AI is almost like having an additional voice in an organization.
Within the context, IEC TC 65 which covers Industrial-process measurement, control and automation, will be another potential group to liaise with for AI and industrial automation.
SC 42 is also planning to collaborate with other external organizations working on AI. At the inaugural meeting, the committee approved a Category A liaison with IEEE with additional future liaisons anticipated.
SC 42 is mandated to serve as the focus and proponent for JTC 1 standardization programme on AI and provide guidance to JTC 1, IEC, and ISO committees developing AI applications. During the meeting, it set up a structure to allow the ecosystem approach that will include:
Foundational standards (Working Group 1)
Given the diversity of AI stakeholders, it is essential to have foundational standards that provide for a framework and common vocabulary. This enables stakeholders of different backgrounds and perspectives to speak the same language and sets the stage for how they and the technology providers and users will interact together. A priority will be the development of the International Standards for AI concepts and terminology ISO/IEC AWI 22989, and Framework for artificial intelligence systems using machine learning ISO/IEC AWI 23053.
Computational approaches and characteristics of AI systems (Study Group 1)
At the heart of AI are the computational approaches and algorithmic techniques that empower the insights provided by AI engines. IT advances, specifically computational power, distributed computing methods and software capability techniques among others, allow for what was science fiction to become science faction. Standardization and best practices in this area are essential if innovation is to occur over open standards. SG 1 will:
Trustworthiness (Study Group 2)
Connected products and services, whether a vehicle, smartphone, medical device or building security system must be safe and secure or no one will want to use them. The same goes for critical infrastructure like power plants or manufacturing sites. Trustworthiness and related areas from a system perspective, such as robustness, resiliency, reliability, accuracy, safety, security, and privacy must be considered from the get-go. Leading industry experts believe that ensuring trustworthiness from the outset is one of the essential aspects to wide-spread adoption of this technology. SG 2 will:
Use cases and applications (Study Group 3)
Use cases are the currency by which SDOs collaborate with each other. As both the focal point of AI’s role as an enabling horizontal technology and in its role as an AI systems integration entity committee tasked with providing guidance to IEC, ISO and JTC 1 committees looking at application areas, it is essential for SC 42 to collaborate with other committees and bring in their use cases.
For example, experts in AI algorithms, who may never have set foot in a factory, will be able to liaise with domain experts in the TCs who come from industry and are able to make the use cases more meaningful, so that the subtleties can be understood – such as the difference between machine learning and neural learning, or how algorithms are trained. This means flagging up that it is not just the algorithms that need correcting, but also the datasets for training. In this way, use cases provided by other committees looking at different vertical application areas can allow SC 42 to consider these technical requirements as it drafts its standards, technical reports and best practices. SG 3 will:
JTC 1 will transfer the work programme for big data (JTC 1/WG 9) to SC 42. Initiated a few years ago, it has two foundational projects for overview and vocabulary and a big data reference architecture (BDRA).
These projects have received tremendous interest from the industry. From a data science perspective, expert participation, use cases and applications, future anticipated work on analytics, and the role of systems integration (working with other ISO, IEC and JTC 1 committees on application areas), the big data work programme lines up well with the initial work programme for SC 42. From an industry practice point of view, it’s hard to imagine applications where one technology is present without the other.
“It stands to reason that AI will be one of the most crucial enabling technologies in our lifetime. JTC 1/SC 42 is looking at the entire AI ecosystem from an IT perspective. Combined with the breadth and depth of application areas covered by IEC and ISO, the resulting standardization efforts will not only be fundamental to practitioners but essential to all stakeholders interested in the deployment of AI in the respective verticals”, Diab concludes.