Listen to the MP3 audio podcast here.
|
AI is everywhere. It is advertised in all social media platforms, TV, magazines, and I have to admit, since I am not a sophisticated user, what I tried so far is between just OK and impressive. For instance, I have been curious about using someone else’s voice in putting together the podcast version of this analysis. So everything that I just said now, an AI voice would say this:….
Meet Daniel. He is just one example of how AI could reshape the voice business, for instance. If you are radio personality or a TV announcer, well, the writing’s on the wall. Personally, I still like the human touch of language, but the transition to the unknown is clearly underway. I remember a lot of people who used to love the old-fashioned print newspaper, eventually managed to transition, after claiming that the world is in trouble because of the internet. The Internet may have displaced a lot of jobs, but it also created probably a lot more. I recall vividly some of the major changes that occurred, occurred rather smoothly and had a lasting effect on us, on me in particular. For instance, with my family living abroad, I gave AT&T and MCI in the 1990s thousands of dollars in long-distance and international calls to my family. Today, that’s no longer the case. I use WhatsApp and done. AI is doing this to us now. It has a major transformative impact and in most cases, we will barely feel it. This reminds me when I set out to teach my 70-year-old mother how to use a smartphone to call me over wifi, It did not take her too long to learn the process, despite being scary in the beginning and some frustration, and now that’s the only way we connect.
AI is doing that and it is doing it at a mind-numbing speed. Its use will be generalized in no time and we may not even know it. It is certainly quickly being deployed in enterprise to the point that businesses are struggling to find experts to help them with their AI projects. More on that in a minute.
Now the question is what’s driving this insanely fast AI adoption? The answer can be found in CIO Magazine’s “The State of the CIO survey” released earlier this year and which lists the top priorities of businesses in terms of their IT budgets. Some 56% reported that they expected their overall IT budgets to increase this year and while AI is nowhere to be found in the list, here are the top 5 areas of focus.
- Increasing operational efficiency: 45%
- Increasing cybersecurity protections: 44%
- Transforming existing business processes: 38%
- Improving the customer experience: 36%
- Improving profitability: 27%
What these five areas have in common is that they are all addressed and tackled by Artificial Intelligence one way or the other, and so because AI has a broad reach to all of these segments of the IT budget, we should expect to see AI everywhere going forward.
So, for those of you who are ITAD companies still wondering about AI, generally the same AI technology used in the greater ITAM sector, that is the greater IT asset management sector, is the same technology stack that you can adopt and deploy specifically in ITAD. The same technology can also be used on both the supply and demand sides, which means this can create opportunities for some interesting new interconnections and integrations between client and provider, that would benefit both the demand and supply sides in terms of efficiencies and cost reduction.
For now, however, it is true that deploying AI in IT Asset Management (ITAM) in the enterprises is a complex process that requires careful planning, execution, and a lot of money. While the difficulty level can vary depending on the specific context and requirements of each organization, there are several common challenges that enterprises may face:
Data Availability and Quality: First, is the availability of data that is of high quality and high reliability, as AI relies heavily on data. If you do not have good quality data, it would be hard to create conditions and outputs that make sense. Getting good data can be a significant challenge, and extremely costly, especially in large enterprises with diverse IT systems and data.
Integration with Existing Systems: Secondly, now if you have the data that you need to feed into an AI platform, linking AI solutions with existing IT systems and workflows can be complex. It can force you to overhaul legacy systems, data migration, and coordination with various stakeholders. And so when you are ready to adopt AI, be prepared to challenge your existing IT platform, and that could be painful.
Resource Allocation: Thirdly, you must have the resources to engineer an AI deployment and an IT transformation. Dedicated resources such as skilled data scientists, AI engineers, and IT professionals, will be needed to think about the process, the design, program, deploy and monitor. Allocating the necessary resources and expertise can be a challenge, especially for organizations with limited budgets or competing priorities.
So what AI appears to be doing in enterprise deployment is to impose change management and new ethical considerations. On change management, organizational changes are likely inevitable unless the scale and scope of the AI project are limited to siloed functions and operations. If it is broader, you will need to re-train employees on acceptance and endorsement, on the new technologies, redefine roles and responsibilities, and face potential resistance to change.
Despite these challenges, deploying AI in ITAM and ITAD would eventually result in significant benefits such as improved efficiency, cost optimization, and enhanced decision-making. The ITAM Review magazine listed a number of benefits for ITAM in a September issue called “The coming together of AI and ITAM.” Broadly speaking, the article lists such benefits as Enhanced Asset Discovery and Inventory Management, where AI-powered tools replace legacy databases for the purpose of scanning an entire network, identifying devices and software, and categorizing assets. AI can be designed to ensure that no critical asset goes unaccounted for. Moreover, AI should be designed to provide detailed reports about asset specifications and configurations, enabling better decision-making and resource allocation.
Think of predictive analytics, which may also be the backbone of predictive pricing so critical to ensuring optimum return from the secondary market. AI can monitor asset performance and usage in real-time, flagging any anomalies or issues as they occur. By leveraging predictive analytics, AI can help identify potential problems before they impact operations. The article also lists such areas as Cost Optimization, Compliance Management, security, etc., that could benefit from AI adoption.
More specifically for ITADs, AI can positively impact the Automated Asset Identification, by scanning and analyzing purchase records and asset inventories, to automatically identify IT assets that are ready for disposition. This streamlines the identification process and ensures accurate tracking of assets. Think of Predictive Analytics for Asset Value, Asset Routing optimization, Data Sanitization and Security, Quality Control and Testing, etc.
In understanding what is being done in the enterprise related to AI, Deloitte recently surveyed 2,620 global business leaders in 13 countries and what jumped at me in terms of impact on critical ITAD functions are the set of tools available for IT operations management, through which end-user companies and ITADs could deploy to manage their IT asset disposition….. These are new platforms known in the tech sector as AIOps (short for AI IT operations management). These are a set of tools that create and enhance intelligent alerting, root cause analysis, anomaly and threat detection, incident auto-remediation, and capacity optimization.
AI also offers many ways to automate time-consuming processes, by removing the potential for human error. Process automation is certainly an area where ITAD companies can leverage AI to design and implement. Several companies have already deployed AI for Financial reporting and accounting where AI improves the reporting efficiency and planning, that would extend to accounting, taxation, cash flow, etc. HR is also a great prospect for the efficient utilization of AI, as in hiring process to screening resumes, matching candidates with best suited roles, etc. In the areas of safety and quality, AI can help automate and streamline these functions to minimize risks to the company.
Now that you have identified where AI may be deployed, you will need to hire people to design, program, deploy and monitor your AI project, and I must warn you that finding the proper talent is likely to be extremely difficult and costly. Sarah White of CIO magazine listed the most sought-after skills on the AI front that are in short supply.
So what skills are companies hiring for exactly? Technicians with Natural language processing (NLP) skills are in high demand as they enable improvements in chatbots, AI assistants, automation, and other tasks.
Technicians with TensorFlow skills. TensorFlow an open-source machine learning framework developed by Google used to build and train machine learning models and neural networks.
Technicians with image processing skills, which would allow the analyzing and processing of images, while also pulling data and information from visuals and text documents, and interpreting or manipulating that data as needed.
Technicians with PyTorch skills. PyTorch was developed by the Facebook AI Research (FAIR) team in 2017 as an open-source machine learning library, a framework that helps organizations build and train deep learning models and neural networks.
Technicians with AI chatbot skills, with companies increasingly using chatbots to lower the burden on human representatives in customer support environments.
Technicians with model tuning skills, who will help fine-tune the settings and parameters for machine learning and deep learning models.
Technicians with Stable Diffusion skills, a deep learning model that produces high-quality artwork and images based off complex and detailed user prompts.
Technicians with Midjourney skills, which is an AI service that generates images using natural language prompts.
Last but not least, you will also need AI content creators for social media and marketing, including experts in creating blog posts, social media posts, graphics, articles, and even videos. Technicians with ChatGPT skills are in high demand because they are the ones who will deploy ChatGPT for content generation, task automation and scripting, translation, on-demand learning, technical support and troubleshooting, editing and proofreading, idea generation, calendar scheduling and management, and more.
The list is big, the functions that can be improved by AI are endless, and so are the challenges.
When you analyze the vast array of challenges surrounding the explosion of AI, where do you even start? We can address ethics but that’s a very complex topic. Something simpler to address may be the issues of intellectual property and regulations. On the first point, there is not much happening on the US copyright law, as stakeholders have not yet involved courts to address copyright litigations. We are already seeing lawsuits here and there against companies like OpenAI on how protected third-party content is used to feed into an AI model in the United States.
In the United States, just as we experienced with regulating electronics recycling, most of the AI regulations are likely to come from states and not necessarily from the federal government for the time being. Just like recycling, AI could evolve as a patchwork of state regulations that could have an impact of its development. California’s governor has been pressuring legislators to address the use of AI in state agencies and departments. He has ordered agencies to produce risk assessment reports on how AI could impact the workings of the state. Oklahoma has a new AI task force to study, evaluate, and develop policy and administrative recommendations for AI deployment. Worried about AI, Maine has already banned the use of AI tools for government, and Washington State issued warnings about it.
The nonprofit Electronic Privacy Information Center said that 10 states have embedded or are about to add AI regulations specifically in their consumer privacy laws, addressing such things as facial recognition, HR tasks, etc.
The European Union is in no better position as well, but Japan appears to have made some inroads there, allowing the use of copyright works for AI training to take place.
So in conclusion, what are we saying?
- ITAD companies can adopt and deploy the same AI technology used in IT Asset Management (ITAM).
- Deploying AI requires careful planning, execution, and resources due to challenges such as data availability and quality, integration with existing systems, and resource allocation.
- AI brings change management and new ethical considerations for enterprises.
- Benefits of deploying AI include improved efficiency, cost optimization, enhanced asset discovery and inventory management, predictive analytics for pricing/resource optimization, compliance management, security, automated asset identification, predictive analytics for asset value/routing optimization, data sanitization and security, quality control and testing.
- AIOps platforms are available for IT operations management to manage IT asset disposition.
- Jobs in demand include technicians with skills in Natural Language Processing (NLP), TensorFlow/PyTorch, image processing, AI chatbots, model tuning/Stable Diffusion/Midjourney and content creation.
