Cutting Through the Hype!
A CEO’s Comparative Insight on GenAI vs DLT Innovations
“Gen AI.. just a new tech hype like DLT!”
“Yea DLT was going to transform everything, AI just sounds the same. Big ideas, no implementation.”
As a strategic tech advisor, I have the opportunity to review and conduct due diligence on companies of various sizes and stages of maturity. My work often involves collaboration with private equity firms, venture capitalists, or company boards, allowing me to gain a comprehensive perspective on how technology influences the strategies and roadmaps of these organizations. This experience has shown me that poor decisions can cost companies significantly, sometimes resulting in more than just financial losses. It’s not uncommon to see executives pushed out of their roles for being overly innovative, too conservative, or for demonstrating either too passive or aggressive a stance. Drawing from my past career as a CXO, I understand that being an executive is particularly challenging in a world where strategies aren’t solely dictated by market forces. Adapting to the rapidly changing landscape is a daunting task.
If you are a CEO, CPO, or CTO who has been burned by the Web3 and crypto hype, you might be inclined to dismiss the current AI revolution as just another passing trend. In fact, many CXOs I meet are investing in AI-related capabilities primarily due to investor pressure. When asked about the depth of their AI capabilities, they often draw parallels to their failed investments in crypto, Web3, and other DLT-related transformations.
As a strong supporter of Web3 and Industry 4.0, I understand their sources of suspicion and frustration. DLT was predicted to transform every aspect of technology and touch every digital product. It was expected to decentralize supply chains, currencies, and digital assets, in conjunction with Industry 4.0’s distributed lines of production, potentially revolutionizing every vertical from banking to sportswear production. But why didn’t it, and why is AI, especially GenAI, unlikely to follow the same path?
There is no single answer to this question, but if I were to summarize it in one sentence, it would be:
“Unlike DLT, GenAI primarily transforms, optimizes, and simplifies the way customers interact with companies’ products and services.”
In fact, companies can see an immediate increase in revenue after utilizing GenAI, without needing to change any aspect of their core product. The CPO can now offer a personalized online sales experience, where the AI sales representative engages in a sophisticated dialogue with the customer, inquiring about their dentist appointment last week before referring them to the latest iPods in the online shop catalog. Behavioral analysis becomes an ongoing process, applying dynamic sentiment analysis based on complex conversations. This eliminates the need for expensive, combined behavioral analytics practices like using clickstream data and purchased social media data about customers’ swiping patterns. For the customer, this could be the first time they experience what our parents did when visiting the same local shop where the salesman knew them, made them feel good, and saved them time by offering exactly what they needed. It’s a kind of stickiness that makes sense.
For the Enterprise Architect, General AI represents an incremental capability and a new integration pattern, without necessitating a complete overhaul of their architecture, from infrastructure to the strategy layer. The AI transformation is not overly complex; the risks are manageable, and the rewards are quickly achievable.
DLT: The holy Pain in the A…
On the opposite side of the spectrum, almost all DLT applications promise optimization of product flows by providing transparency and simplicity in transactions, in exchange for a complete overhaul of the core product specifications, enterprise architecture, and often delivering a complex or unconventional experience for the user. Transactions in an information system are the vehicles through which the state of the system transfers. They are fundamental and can be compared to the foundational central pillars of a building. Changing their nature touches every aspect of the business, every compliance policy, security protocol, and often the user interaction with the system. The rewards associated with these risky and expensive transformations are often far in the future, ethereal in nature, or involve replacing an existing and functioning control structure without an obvious immediate benefit for the customers or business stakeholders. Of course, they are revolutionary in nature, but how often in history do we encounter revolutions working in favor of the revolutionaries themselves?
Comparing General AI with DLT transformation is almost like presenting two plans on a hotel owner’s desk for next year’s renovation. One plan spends money on training staff, upgrading the kitchen, and remodeling the rooms to gain an additional star, compared to replacing all walls, floors, and doors of the engine room and management offices with glass to provide transparency of the hotel’s affairs to the guests.
Of course, there are many use cases where the examples above may not apply, but as an executive, rest assured that soon your customers will not be interacting with your products unless it’s in conjunction with or through General AI.
In our next publication, we will dive deeper into LLM-based architectures that provide personalized experiences to customers, utilizing various techniques such as RAG, data-driven prompt engineering, and vector database partitioning. We will also explore OpenAI’s current capabilities to interact with your product inventories and recommend products based on dynamic behavior and sentiment analysis, taking into account both the current session and historical context.
So, stay tuned, and thank you for reading the latest from The Good CTO.
