A simplified spectrum

Considering the strategic, structural, and cultural implications of generative AI can be challenging. I have found that viewing generative AI on various levels of sophistication and complexity helps to gain better clarity. This “spectrum” can in turn be simplified into four perspectives and seen as a staircase, where one level if advanced enough leads to the other. For each step the power of generative AI grows, but so do the implications of what we may have to do to make it succeed and our risks from miss-managing said actions. Each step can also help create a shared clarity by being a form of “classification” aid.

Nevertheless, before continuing it is critical to highlight two things. Firstly, this is a simplification to help see the “trees of the forest”. Secondly, even though seemingly on an “IT” level, the real implications are often in strategy, organization elements and culture.

The four perspectives are:

• Simple digital transformation/tool continuation
• Next level computer interface
• The new co-worker
• Part of the “meta” transformation

This reflection will go through each perspective above including the pros and cons and finish with concrete examples.

Simple digital transformation/tool continuation

The most direct way to approach generative AI is to simply see it as a continuation of the digital transformation that has been taking place over the last decades. In other words, we are going to get a better word processor, calculator, presentation tool, collaboration tools, search engine and so forth.

• Pros: This angle is a great starting point for having a simple reflection exercise and helps you identify when AI is a direct extension of your current situation
• Cons: It misses most of what generative AI can, or should be enabled, to do. Also, be sure that you are not seeing something as a direct tool “2.0” that really has much more power for gains and damages.

Next level computer interface

Another way to look at generative AI as a continuation of the human computer interface evolution. We have over time moved from simple interfaces to text direct on screen to graphical interfaces, and even reasonably competent voice and gesture commands. Each step has brought us closer to being “one” with IT technology; offering such capability that it can even remove previously necessary human middlemen.

• Pros: This angle offers the ability to see places where we can create much more efficient and effective interactions. Be it the next generation intranet or the more direct human machine support experience. It also provides some “sanity” check to if a task may fit generative AI without overextending it.
• Cons: It does have the challenge of missing the foundation perspective more easily seen by the digital transformation perspective and still misses when generative AI goes beyond simply improving the interface.

The new co-worker

In this article I was referring to generative AI in a specific context as essentially a co-worker. This is both to help us see this technology in a different light, but also because it can start act as if it was a “co-worker”. This is when the generative AI, through its ability to take in input (text, voice etc.) and generate a human like response combined with its other capabilities such as searching and compiling massive amounts of information, acts as a co-worker. For example, a generative AI working with the information and knowledge available either externally or internally can start doing the same role as various parts of your talent pool, be it managing a calendar, helping during or after meetings or going beyond searching for support knowledge by trying to train the person asking the questions.

• Pros: This perspective as a general idea better illustrates how sophisticated generative AI requires that one take very seriously the structural and cultural aspects inherent in viewing a machine as a co-worker. If the machine is your executive assistant, your support service personnel or personal artist, you also know the risk of giving too big challenges or not making sure it can deliver. It also covers some extremely powerful scenarios where new capabilities outside the human realm emerge, such as handling massive pools of knowledge, blistering fast artistic drawings and another form of creative thinking/divergence of “thought”.
• Cons: Firstly, it is apparent, but if you do apply generative AI as a co-worker and not consider it holistically you do take some great risks. Also, one must remember that the field is continuously in flux and strategic foresight is paramount to ensure you do capture new areas for value gain. Finally, one must also ironically not over-extend as there clearly are areas where generative AI is not up to the task, even if the results we see may appear convincing.

Part of the “meta” transformation.

We now progress from the area of generative AI into a fuller spectrum of interlocking developments. Instead, we see the full next digital transformation, which I prefer to refer to as a “meta” transformation. This is where generative AI interconnects with our hybridization of work across time, place and intelligence and share the room with other technologies such as augmented reality.

• Pros: It is in this cross section of ongoing technological trends and their realization from strategy to culture where one can see the full scope of value. Some truly relevant use cases also cannot likely be considered without this perspective.
• Cons: It is also here where one most easily runs the risk of missing the trees for the forest and potentially overextending. It is complex to consider generative AI alone, hence the demand on your strategic foresight, planning and execution naturally grows. One key example of this is how we now may move into multiple “new” fields at the same time, such as having to introduce multiple interlinked technologies and ways of working to workers who may lack much experience with any of them.

Putting the perspectives together to provide clarity, direction, and action

If you look at your current landscape from the perspectives in this article you should be able to gain valuable insight into how generative AI can help you. The perspectives can also be illustrated through questions and examples, as done below:

• Scope: Is it about AI or more? For example, if you detect that you would require the “eyes” and “ears” of AI to reach into your factory and you need to provide the AI “hands” to point, you are likely also going into a journey of augmented reality. While you do, do not forget that you are potentially introducing multiple technologies and ways of working that may also involve topics such as digital know-how, feasibility, and ethics.
• Talent replacement: Are there areas today where your talent essentially is just an interface? If yes, can it be automated? For example, if your first line of support essentially acts as a sophisticated interface there may be room for a new digital co-worker. Do not forget however that replacing all humans with one machine does put a lot of importance on that machine working as intended.
• Surfaces to knowledge: Where are your greatest problems of discoverability and findability? For example, your intranet news and your search engines results may be ready for closer human machine integration. However, making things easier to discover or find also puts greater emphasis on that your applied information security classification really does work.
• Enhancement: Look at your current tool suite. More providers are letting you utilize generative AI as a part of their already existing tool. However, as you do contemplate if the nature of the tool has changed more in line with the other perspectives.

My hope is that one or multiple of these perspectives will help you see how generative AI is or can impact your organization better and enable you to better leverage that value. For additional information on how to help enable AI in a digital knowledge landscape context, I highly recommend my previous articles.