This week AJ had the opportunity to speak with Alex Cojocaru from Licenseware to look beyond the hype of AI. What is AI, what does it mean for ITAM Professionals, and where will we see it being used in the next few years?
The conversation was recorded in this podcast:
You can download the podcast wherever you prefer to listen, on Spotify, SoundCloud and Apple Podcasts.
Alex was also kind enough to prepare some words on the topic, which you can read below.
It is fair to say the ITAM community approaches emerging technologies like generative AI with curiosity and scepticism. While some of us see the potential, others question the immediate applicability and utility of AI. Let’s untangle this sticky topic and explore how generative AI could be a transformative force in ITAM.
The automatic door doesn’t replace the doorman. The automated door may handle the basic task of opening and closing, but it can’t replace the multifaceted roles of a doorman, such as providing a sense of security, offering personalized service, and even acting as an informal concierge. In essence, technology can automate functionality but often falls short in replicating the nuanced human interactions that contribute to a building’s community and character.
The current AI tech is like a bicycle for the mind, primarily if we’re referring to generative AI (e.g., ChatGPT). It amplifies ITAM professionals’ capabilities and makes them 10x ITAM Managers. It’s worth mentioning that, like any tool, it needs mastery and understanding. Continuing the bicycle analogy, some people can ride fast on their bikes, while others can do wheelies and cool backflips. While some people can barely ride it at all. They might even say it’s not all that anyway. Also, context is essential. You wouldn’t ride your bike on ice or a dune. The bike is useful in a particular context, and so is generative AI.
“AI” is often thrown around loosely, sometimes serving more as a buzzword than an accurate descriptor of the technology in question. While AI aims to create machines capable of general human-like intelligence, most of what we see today is Narrow AI, or Artificial Narrow Intelligence (ANI). These systems excel at specific tasks but lack the broader contextual understanding that would enable them to perform any intellectual task that a human can do.
Machine learning, a sub-branch of AI, focuses on computer systems learning from data. It’s statistics married with computer science on steroids. Alongside neural networks and deep learning, to name a few, these technologies contribute to the broader AI goal of reaching Artificial General Intelligence (AGI) and beyond. These technologies are specialized and limited in scope. So, it’s crucial to differentiate between these various subsets when discussing “AI” and avoid misleading stakeholders and the public about existing technologies’ current capabilities and limitations.
Generative AI is reshaping the landscape of content creation and automated systems. Beyond mere prediction, if used well, these algorithms can produce entirely new, coherent, and contextually relevant content. They can be fine-tuned for specific tasks such as chatbots, automated journalism, and even generating code. However, it’s essential to understand that while the output may seem intelligent, it’s ultimately based on patterns in the data it was trained on. The system doesn’t “understand” the content it generates in the way a human does; it mimics the statistical properties of its training data to produce the most statistically plausible results. While Generative AI is powerful, its capabilities should be contextualized within its design and training boundaries.
The reality is that generative AI systems like ChatGPT are not just technological novelties; they’re functional, practical tools that are seeing increasing integration into various industries. The hype isn’t entirely unfounded—thanks to the maturity of the technology, interacting with these systems is as simple as typing on a keyboard, making them highly accessible. Their utility for certain use cases has far exceeded initial expectations.
However, the challenge resides in leveraging these tools effectively. For the untrained consumer or business professional, the saturated market of AI offerings, compounded by the marketing buzz, creates a confusing landscape that makes it difficult to select the most suitable tool. A Harvard study of 758 consultants sheds light on the utility of AI in professional settings, revealing that AI can substantially enhance work quality—by up to 40%. The study found that lower-performing consultants benefitted the most, experiencing a 43% improvement in speed and quality, while high performers saw a more modest 17% gain. This suggests that AI can serve as a democratizing force in the workplace, levelling the playing field.
In practical terms, people typically engage with Generative AI in two distinct personas:
While Generative AI offers immense potential, its effective application hinges on nuanced understanding and strategic use. Whether one chooses the Centaur or Cyborg persona, the key is to employ AI as a complementary tool rather than a complete replacement for human intelligence and expertise.
AI excels in contract management, data analysis, and knowledge management tasks. Imagine a tool that reads through your vendor contracts, extracting essential dates and clauses while acting as an internal advanced search engine.
Roles adjacent to SAM, such as sales and procurement, could immensely benefit from a specialized licensing chatbot. This bot could offer real-time information on vendor licensing, streamlining several operational processes. For this to work, the model is trained only on official data, and each answer must paired with the original source reference.
As these AI tools advance, they could serve as co-pilots, providing data-backed, high-level insights to decision-makers. A CIO could ask about asset utilization or upcoming renewals and get instant, strategic responses, shifting AI’s role from a tactical assistant to a strategic advisor.
ChatGPT, the first mover and most popular solution in the space is a black box. And it’s known to hallucinate a lot. In simple terms, you can’t trace the answers to a source(s), and it may produce a solution that is not correct, entirely or partially made up. The scary part is that these answers sound legitimate. So, the only way to know is to fact-check all the answers. As you can expect, this setup does not inspire trust in ITAM, where we rely on correct information and data.
On the bright side, other newer systems and models can provide sources and do not hallucinate. They admit that they don’t know the answer. I expect to see this trend taking off in ITAM soon.
Over the next few years, I foresee AI evolving as an integral part of ITAM solutions. Its capabilities will be refined, and as professionals become more adept at using it, we can expect to see a surge in AI-powered ITAM tools and services offering strategic advantages.
There are most likely numerous other applications that haven’t yet been discovered, so I’m very excited for the next decade and the opportunities it will bring.
Alex will be delivering a session all about this topic on Wednesday 15th November at 11.30. Sign up to Wisdom APAC 2023 to hear more.