An old technology with a fresh coat of paint has taken the world by storm, and the hype is justified this time! Well, maybe. And it is much more than a fresh coat of paint. We may be seeing an AI Renaissance, possibly meeting the expectations set decades ago where computers did everything, and a hard day of work is when you had to push three buttons. But this technology dropped seemingly out of nowhere, and the scramble has begun to understand its impact on AI and the entire technology industry. One thing is for sure, ITAM must monitor this technology to assess the organization's interest and how to prepare for its adoption. If some extreme opinions play out, this technology will disrupt more than just your IT standards. It could disrupt your entire ITAM program. Let's learn more about ChatGPT, what it currently does, and its potential impact on you, your organization, and your industry.
OpenAI's ChatGPT [1] is a bright shiny new object that has caught the attention of just about everyone. ChatGPT is yet another application of Artificial Intelligence (AI). Still, this time appears to have taken an enormous leap forward to produce well-thought-out and impressive answers to your questions as if a human being wrote the explanation.
So that's my description of ChatGPT, but what does ChatGPT say about itself? I asked, and here's the answer:
ChatGPT's answer:
ChatGPT is a pre-trained language model developed by OpenAI. It is a transformer-based neural network model that has been trained on a large corpus of text data to generate human-like text. The model uses deep learning techniques to understand the context and generate relevant and coherent responses to natural language inputs, making it a powerful tool for a variety of language-based tasks, including text generation, question answering, and conversation. In a conversational setting, ChatGPT can be used to build chatbots and virtual assistants that can understand and respond to user queries in a human-like manner.
That's a lot of techno-speak! But ChatGPT is gaining so much momentum that Microsoft has reportedly invested billions of dollars in OpenAI [2]. And others have taken notice, including Google [3]. Does this mean we are done writing mundane reports? Can ChatGPT write my status reports? Okay, let's be an adult for just a second. If I asked ChatGPT to write something, do I own the copyright on ChatGPT's work? That's a great question!
Let's start with this from the U.S. Copyright Office:
"Works are original when they are independently created by a human author [4]"
Uh-oh. It doesn't sound like you can claim ChatGPT's work as your own. Not unless there is a man behind a ChatGPT curtain that we are not supposed to pay attention [5] to, and then he is the copyright holder unless he transfers that right. Copyright may not be a concern if you use ChatGPT as an assistant. Next question, can you claim credit for ChatGPT generated content? Or do you have to cite ChatGPT's work? We’ll come back to this a little later.
While all of this is certainly interesting, I am sure you are asking yourself, "Why does this matter to Software Asset Managers and the ITAM program?". Let’s dive in.
For those organizations with a solid SAM program, adding a new application to the approved catalog requires investigations, reviews, negotiations, and approvals before it is accepted as part of your IT standards. But what if a new application could be accessed on company-issued technology without the ability to track it through standard methods? This raises more questions.
How would you know if it were? ChatGPT is accessed from a web browser making discovery a bit more challenging and requiring URL monitoring. It can also be accessed through an API which means you will need to know if software developers are using the API. The question is, should ChatGPT be part of your IT standards? And for whom? Let's look at three of the main drivers behind the decision of choosing to manage ChatGPT as a cloud service asset:
It is free to use the ChatGPT client for now. Relying too much on ChatGPT may become costly and will need justification in the future for its use.
ChatGPT API is not free and must be paid for based on usage [6] as of this writing. The pay-as-you-go model may result in high costs.
How is the information you provide stored, where is it stored, and for how long? How are your identity and usage protected? Can the APIs be used maliciously?
ChatGPT has terms and conditions for use. Users must be aware of these terms and conditions, and SAMs must monitor the terms for any changes that can be made without notice.
Until now, we have generalized ChatGPT's output as just content, but what kind of content can ChatGPT generate? Would you believe software?
The following is an example of code that ChatGPT generated.
ChatGPT's answer:
Here's an example of a bubble sort function to sort an array of 20 integers in ascending order in C:
The bubbleSort function takes an array arr and its size n as arguments and sorts the array in ascending order using the bubble sort algorithm.
The algorithm works by repeatedly comparing adjacent elements and swapping them if they are in the wrong order. This process is repeated until the array is fully sorted.
When it comes to code generation, OpenAI clearly warns you the code generated may not be correct or optimized for its intended purpose.
SAMs manage software according to the entitlements defined by copyright laws and licensing agreements. SAMs are also interested in where the software comes from to ensure the software is genuine and the software does not introduce any security risks. Publisher-generated software or organization IT-generated software can include software from other manufacturers.
As mentioned earlier, a work must be generated by a human to be awarded copyright protection. Does this mean the ChatGPT generated software is free to use? No, due to the terms of use by OpenAI and thus the challenge for SAMs. A SAM's organization should not use software that violates the rights as set by copyright law and the licensing agreement. The use of ChatGPT generated software can easily create a similar non-compliance risk.
Perhaps you are thinking right now, oh yay, more complexity. Thanks, Keith! Regardless, I found this exercise very impressive, even if it were for a simple algorithm. So is the hype justified, and has AI taken the next giant leap? Has Skynet [7] been made self-aware?
Gartner has documented the hype cycle [8] because we have all seen the hype when a new technology is introduced. At some point, the hype subsides, and the latest technology reveals its true capabilities; the market moves on to the next biggest thing, or the market begins to adopt the technology, but this time with a clear vision of what the technology can do.
Companies are investing heavily in AI solutions for numerous markets, such as healthcare, life sciences, and cyber security. While the predictions are that ChatGPT will replace Google search, programmers, writers, and even Software Asset Managers, it is highly doubtful given what ChatGPT is – a tool to help humans work more effectively and efficiently. With that said, I asked ChatGPT the following questions.
ChatGPT's answer:
IT asset management is the process of tracking, maintaining, updating, and disposing of an organization's hardware, software, and other technology assets. The goal of IT asset management is to optimize the value, cost, and risk of an organization's technology assets. This includes ensuring that the organization's technology assets are used in an effective and efficient manner, that they are secure, that they are compliant with legal and regulatory requirements, and that they are disposed of properly at the end of their life cycle.
IT asset management typically involves activities such as tracking the inventory of hardware and software assets, managing software licenses, monitoring the use and performance of assets, and tracking changes in technology assets over time. It also involves making decisions about the acquisition, deployment, and disposal of technology assets based on the organization's goals, budget, and resource constraints.
Not a wrong definition of ITAM, but it lacks acknowledging ITAM is a program that is a horizontal function, crossing the entire organization. Now let's look at the next question.
ChatGPT's answer: Software Asset Management (SAM) is the process of managing, organizing, and maintaining an organization's software assets throughout their lifecycle. This includes software licensing, deployment, usage, and retirement. The goal of SAM is to optimize the value of software assets while minimizing the risks and costs associated with their use.
SAM involves maintaining an accurate and up-to-date inventory of all software assets, including information about their licenses, usage, and expiration dates. This information is used to ensure compliance with licensing agreements and to avoid the potential consequences of using unlicensed software. SAM also includes processes for the efficient deployment and distribution of software, as well as tracking usage and retiring software that is no longer needed.
Overall, SAM helps organizations to better manage their software assets, reduce costs, minimize risks, and ensure compliance with licensing and other legal requirements. It is a critical aspect of IT management and should be integrated into the overall IT strategy and governance of an organization.
Again, this is an excellent definition but is missing some critical points. One point is the importance of acquiring software where those organizations with buying power can negotiate the terms and conditions of the software license. Another point is an accurate software inventory is required for patch management and cyber security.
These two examples demonstrate the surprising results that ChatGPT can generate but also its shortcomings. What should concern SAMs is if people use these definitions within your organization to understand what you do. That is not ChatGPT's job; it's your job.
The hype surrounding ChatGPT may be justified, if not for ChatGPT, but for AI in general. The immediate impact for SAMs is to monitor their organization's adoption of AI, use of rogue AI software, and any changes to IT's roadmap to include AI.
Additionally, SAMs must assess the current and ongoing risks of using ChatGPT. These risks include:
We may be witnessing a historical event in technology similar to the cloud, the iPhone, and the world wide web. The Software Asset Manager must monitor emerging AI solutions and if your organization will use them. So much of SAM may appear to be managing what is currently being used, however, being proactive before new technology is adopted pays dividends once that technology ends up on your network.
Meet with your IT department, your IT cyber security department, and your business leaders who see IT not as a necessary evil but as a competitive advantage and get ahead of the game.
[3] https://blog.google/technology/ai/bard-google-ai-search-updates/
[4] https://www.copyright.gov/what-is-copyright/
[5] A reference to the movie “The Wizard of Oz” - https://www.youtube.com/watch?v=NZR64EF3OpA
[6] https://openai.com/api/pricing/
Our expertise enables individuals to advance their ITAM program for the future by providing next level IT Asset Management best practices knowledge. These practices create a symbiotic relationship between ITAM and departments such as IT Security, IT, Finance, and HR by working in tandem to provide heightened information quality which significantly reduces risks, creates greater financial benefits, further enhances compliance, and increases efficiencies.
Copyright ©2021-23 ITAM IQ | Akron, Ohio USA | All Rights Reserved