Artificial Intelligence and Machine Learning and the Effect on the Future of Work
On the 10th of March, the second Tech Tuesday took place at the Urban Sciences Building, Newcastle University School of Computing. The topic of the event was Artificial Intelligence and Machine Learning and their Effect on the Future of Work. The event welcomed main speakers, including Dr Graham Morgan, the Director of Networked and Ubiquitous Systems Engineering (NUSE) Group (School of Computing), Charles Sellers, the CEO of the Innovation and Technology Group and the founder of the Digital Growth Hub, and Steve Cammish, VP Edge Solutions for AI & IoT at ADLINK. They addressed the programme of the event by giving insights into what artificial intelligence (AI) and machine learning (ML) are, discussing what these technologies can do for organisations, where they can be attained and what costs to expect.
Dr Graham Morgan started the event by introducing into machine learning. He explained the difference between traditional programmes and machine learning. While in traditional programming, the programmer codes the behaviour of the programme, machine learning works based on learnt approximation. Based on any input data, ML can be used 1) to find data classification, semi-classifications and regression, 2) to discover the structure of patterns in the data and identify the abnormalities in it, 3) to improve and see what can be done differently. ML can help deal with the data, which is large and varied, which requires a different way to query it and if different codes are required to answer different questions. Further, throughout the presentation, Dr Morgan mentioned statistical approached and tools that are required to structure data. The presentation proceeded by highlighting some challenges that humans face with the output that ML produces. The concerns are related to output validity, fault-tolerance ambiguity, provenance issues, systems’ proneness to attacks and human-in-the-look propagations. The challenge is not how to build anonymous systems using ML techniques in fully connected infrastructures, but how to fit it in with humans. In conclusion, Dr Morgan provided examples of the projects that had been done. He pointed out that AI gives prolific research and business opportunities as we are just at the dawn of AI development.
The second speaker was Charles Sellers, who started off by briefly reminding about the ultimate goal of the Tech Tuesday events in sharing knowledge and expertise to address emerging technologies. The presentation by Charles Sellers was delivered to provide a short and comprehensive introduction into AI, the value of AI, ML and predictive analytics, as well as the opportunities AI provides for controlling, monitoring and the automation of tasks. AI was explained in the context of a human to explain to non-tech savvy people how AI works (i.e. symbolic-based and data-based functioning). The presentation covered the functions of machine learning in processing data for classifying and predicting things. Also, three ways of working with algorithms, such as supervised learning, non-supervised learning and reinforcement learning, were touched upon. The second part of the presentation provided the food for thought about the future of AI and the way how it impacts the global workforce. Charles Sellers recapped some facts, statistics and opinions published by consulting companies on AI, which helped shape the picture of AI reality. For example, although AI spending is increasing, there are potential but key challenges that need to be overcome such as, dramatically low application of technology beyond the pilot stage, high failure of start-up companies, duplicated AI deployment, a lack of skilled staff and unrealistic expectations. The opportunities and challenged of AI development in Newcastle were also mentioned. Drawing on predictions about the labour market, Sellers concluded that AI is affecting the future of work, in terms of how we work, where we work and the skills we need to work. Against the backdrop of 75 mln displaced jobs, 133 mln new jobs will be created, having a more virtual and flexible nature. AI will redefine the demographic profile of workers, by stretching working age limits. Also, the majority of global CEOs expect that the most innovation in the future will be co-developed with partners outside their organisation. Given what was presented, the speed of the introduction of AI in the global workforce and the scale of its impact cannot be fully apprehended now.
The third speaker was Steve Cammish, whose presentation was focused on AI Machine Vision and Machine Health. He provided an insight into the value that Ai-based technology brings to organisations by filling the skill gap, enabling new opportunities for business and making the processes easier to handle. Based on the examples of their products, he explained how IoT and AI can deliver business outcomes and secure higher ROI. The technology can optimise operations, improve quality, provide predictive maintenance, develop on-demand services and create new business models. Edge computing represents cost-effective, reliable, real-time and secure applications. Steve Cammish provided examples of AI-based solutions that ADLINK produced for different industry sectors, such as smart pallets. Enabled by smart cameras and deep learning edge IoT software, smart pallets are aimed to replace handheld packing, palletising and inventory of items. Such a solution may bring staggering ROI due to increased profitability, reduced inventory and packing errors, increased productivity and a safer work environment. Among other solutions is Machine Condition Monitoring Edge Platform, which is used in manufacturing for collecting data 24/7 for rotating machinery and equipment. The remaining part of the presentation was about AI development at ADLINK. The company uses the approach, whereby they reuse and combine software and hardware components for building completely new solutions for industries. Steve Cammish explained that digital experiments in their company are the circular cycle consisting of ideation, value identification, prototyping, value confirmation and productising. However, one of the most important aspects of the development of AI is to team up with the right partners to integrate capabilities and built new solutions. The presentation concluded with the introduction of a new product, called VIZI-AI. It is an industrial machine vision solution, which can be connected to different image capture devices. It can be used to improve machine learning models to harness insight from vision data and optimise operational decision-making.
