📌 The complexity conundrum (is the cloud a fragile giant?), guest post: Tony Fish
AI case studies: August 2024 | How AI is transforming the world?
New email: 📌 The complexity conundrum: Is the cloud a fragile giant?
In the last article, I asked the question, "Is the centralized cloud infrastructure model a ticking time bomb?" I provided the contrarian views of yes and no.
However, this time, Tony Fish wants to add a layer that provides an alternative answer: complexity.No one person knows how to make a pencil
Imagine holding a simple pencil in your hand. It seems straightforward enough, But the journey to create this humble tool is astonishingly intricate. It is widely said that no single person alive knows how to make a pencil from scratch because it requires the collective knowledge of countless individuals.
First, consider the wood. Trees must be grown, harvested, and cut into precise slats. This process involves loggers, foresters, and sawmill workers. The graphite core, often called the "lead," is another complicated story. Graphite miners extract the raw material, which is then purified and mixed with clay to achieve the perfect consistency and durability. This demands time, innovation, and experimentation to get it right.
Then there’s the metal ferrule—the small piece that holds the eraser. This requires miners, metalworkers, and machinists. The eraser itself is a blend of rubber, sulfur, and other chemicals, requiring chemists and factory workers to produce.
Even the paint that coats the pencil involves its own set of specialists, including chemical engineers and industrial designers. Then, let's not forget the packaging, marketing, and delivery to get it into your hands.
Each component travels through various stages of manufacturing and distribution, with countless hands contributing along the way. The pencil, a seemingly simple object, is a testament to the interconnectedness and specialisation of human knowledge and labour. No single person can claim to know every step, making it a true marvel of collaborative effort.
What to expect?
Networks are like pencils
Your mobile phone appears more complicated but so simple. However, starting with the network. It involves a vast infrastructure of cell towers, fiber-optic cables, and satellites. Engineers design and build these components while construction workers install them. Network technicians maintain and troubleshoot the system, ensuring seamless connectivity. Software developers create the applications and protocols that allow data to be transmitted and received efficiently.
Now, consider the mobile phone itself. The sleek device in your hand is a marvel of engineering. Inside, it contains a microprocessor, which is designed by highly specialized computer scientists and manufactured in high-tech facilities. The screen, a sophisticated blend of materials and technology, is produced by another set of experts. The battery, a crucial component, involves chemists and engineers who work to optimize its performance and safety.
Each part of the phone—the camera, the antenna, the casing—requires its own set of specialists. From miners extracting raw materials to assembly line workers piecing everything together and software engineers writing the code that makes it all function, the creation of a mobile phone is a collective effort.
No single person can claim to know every step involved in creating a mobile phone or building a network. These technologies are a testament to the intricate web of human knowledge and collaboration, making everyday communication possible.
Indeed - we still argue about many aspects of how to make it work, especially deep in the weeds of the dark arts of antenna technology, and if IP networks should be synchronous or asynchronous.
Complexity provides two equally valid viewpoints
There is agreement about complexity and risk but complexity creates bias towards different outcomes.
It is so complex that no one person knows how to make it fall over or fail. Stability is achieved because of the vast number of skills and components. Interconnected creates a grid of interdependence. Time reduces risk and increases confidence.
It is so complex, and there are so many unknown weaknesses we cannot fathom or know what questions to ask to help find the vulnerabilities, and therefore we are always open to an unknown catastrophic failure. Time increases risk and reduces confidence.
These are equally valid viewpoints, and the evidence is there to support both. For example, mobile phones and Banking work 99.999% of the time, and yes, it is frustrating when they go wrong. Then there is Crowdstrike and other major outages from vulnerabilities, which showcase that it is all on a fragile knife edge.
History, experience, knowledge, questions, and skills will influence the viewpoints held by different members of the leadership team.
Our story is about AI and the implications we face created by complexity.
Let's start with the underlying algorithms for AI. Mathematicians and computer scientists develop the theories and models that form the foundation of AI. These models require extensive research, experimentation, and refinement to function effectively. Bias in math, thinking, and coding creates different problems we cannot fathom right now.
Next, consider the data. AI systems need vast amounts of data to learn and make accurate predictions.
Data scientists and engineers collect, clean, and preprocess this data, ensuring it is suitable for training AI models. This process involves expertise in statistics, domain knowledge, and meticulous attention to detail.
However, we find data and unintended consequences difficult to comprehend and talk about.
“We find data and unintended consequences difficult to comprehend and talk about.”
Then there's the hardware. Running AI algorithms requires powerful computers with specialised hardware like GPUs and TPUs. Engineers design and build these components, while technicians maintain the data centres where they are housed. Electrical and mechanical engineers ensure the infrastructure can handle the immense computational load.
Software engineers play a crucial role in implementing AI models and writing the code that enables them to run efficiently. They collaborate with user interface designers to create intuitive and user-friendly applications.
This is why diversity matters
This started by reflecting on the question, "Is the centralised cloud infrastructure model a ticking time bomb?" Last week, we looked at the contrarian views of yes and no, and now, about the layer of complexity.
However, this is a critical part that gets missed when we get deep and dirty in the weeds of code, data, and technology—diversity matters.
Diversity allows us to create the complex, think about the complex, and question the complex.
Diversity allows us to work together to recognise, find, and illuminate single points of failure. Diversity makes us more immune to group thinking, risks we cannot see, and poor decisions.
This is a critical part that gets missed when we get deep and dirty in the weeds of code, data, and technology—diversity matters.
In an age of complexity and uncertainty, we need diversity and integration because it is the key to innovation, resilience, and sustainable success.
It empowers us to leverage a wide range of perspectives and experiences, fostering creativity and enabling us to tackle challenges more effectively.
Embracing diversity not only enriches our problem-solving capabilities but also enhances our ability to adapt to rapidly changing environments.
By integrating diverse viewpoints, we build stronger, more adaptable systems that can thrive in the face of complexity and uncertainty.
Ultimately, diversity is not just a beneficial attribute; it is a fundamental necessity for driving progress and achieving excellence in our endeavours.
Explore more
Share you thoughts
How do you think artificial intelligence is transforming the world?
Please take a moment to comment and share your thoughts.
📌 AI case studies
You are receiving this email because you signed up for Wild Intelligence by Yael Rozencwajg. Thank you for being so interested in our newsletter!
AI case studies are part of Wild Intelligence, approaches and strategies.
We share tips to help you lead, launch and grow your sustainable enterprise.
Become a premium member, and get our tools to start building your AI based enterprise.