Three Sigma: Emerging Tech
Edition 3 - Material Sciences, Brain Health, Sora by OpenAI, CEO types
Co-Authored by Nirlep Adhikari and Julian Fayad
Graphene: A Marvel of Material Science
Graphene, often hailed as a "wonder material," is a single atom-thin sheet of carbon atoms arranged in a two-dimensional honeycomb lattice. It is the first 2 dimensional material ever discovered, and its atomic structure gives graphene its remarkable properties, making it one of the most promising materials discovered in the 21st century and it could replace silicone in computer electronics. We are also pretty excited about Borophene, but that is for another edition.
Strength and Flexibility: Graphene's extraordinary strength comes from the covalent bonds between its carbon atoms, which are among the strongest in nature. Despite its strength, graphene is incredibly flexible and can be stretched up to 25% of its original length without breaking. Graphene is up to 200x stronger than the equivalent weight of steel. It has 2x the stopping power of Kevlar on a bulletproof vest, and 10x the stopping power compared to steel plated bullet proof vests.
Electrical and Thermal Conductivity: Graphene's electrons move with almost no resistance, this is because when the atoms form a covalent bond, there is always a valence electron free or available, making it an excellent electrical conductor. This property is pivotal for potential applications in electronics, where graphene could lead to faster, more efficient devices. You could see batteries able to be recharged in as little as 5 seconds. Additionally, graphene's thermal conductivity is superior to all other known materials, making it ideal for heat management in electronics and other applications. Check out this cool video below where the heat from a human hand transfers to a blade lined with graphene to cut through ice like it is knife cutting through a cold stick of butter.
Potential Applications: Beyond replacing silicon in electronics, graphene's unique properties could revolutionise various sectors. In energy storage, graphene-based batteries could charge much faster than traditional ones. In materials science, adding graphene to concrete or plastics can create composites that are lightweight yet significantly stronger and more durable.
Challenges and Future Prospects: Despite its potential, mass-producing graphene in a form that retains its extraordinary properties is challenging. Researchers are exploring chemical vapour deposition (CVD) and other methods to scale up production affordably. The future of graphene lies in overcoming these technical hurdles to unlock its full potential in various industries.
Metamaterials: Engineering Waves and Beyond
Metamaterials are engineered structures, or hybridised materials designed to interact with electromagnetic waves, light waves, sound, and other physical phenomena in ways not found in nature. By manipulating the structure of materials at the microscopic level, new materials with properties that can bend light, sound, and more can be produced. The blending of multiple materials can take the best attributes of each material and combine them.
Cloaking / Invisibility: Metamaterials owe their unique abilities to their structure rather than their composition. These materials can bend electromagnetic waves, including visible light, around objects, effectively making them invisible — a principle behind the theoretical "invisibility cloak."
Sound Manipulation: Acoustic metamaterials can control sound waves by bending, shaping, or focusing them, leading to potential applications in noise reduction, improved acoustic devices, and even sonic levitation.
Refractive Index Control: One of the most intriguing aspects of metamaterials is their ability to achieve a negative refractive index, a property not found in naturally occurring materials. This allows for the development of superlenses that can focus light beyond the diffraction limit of conventional lenses, opening new avenues in microscopy and imaging.
Challenges in Development: The fabrication of metamaterials is complex and often requires precise nano-scale engineering. Currently, the production of metamaterials is expensive and time-consuming, limiting their widespread application.
Future Directions: The ongoing research aims to simplify the production of metamaterials and explore their potential in various fields, including telecommunications, medical imaging, and energy harvesting. As the technology matures, we may see metamaterials become integral to a wide range of innovations, from stealth technology to revolutionising optical devices.
Material Sciences: The Next Frontier
Graphene and metamaterials represent the cutting edge of materials science, offering a glimpse into a future where the boundaries of what's possible are continually expanded. As research progresses, these materials could redefine industries, technology, and our everyday lives. Some examples of applications include:
Rapid-Charging Batteries: Imagine batteries that charge in just five seconds, dramatically reducing waiting times for devices to power up.
Smart Windows: Graphene-coated glass could allow windows to switch between radiative cooling and solar heating modes, optimising energy use in homes.
Construction Innovation: When mixed with concrete, graphene contributes to creating materials that are lighter yet stronger, promising advancements in construction technology.
Superior Conductivity: Graphene's exceptional electrical and thermal conductivity, combined with its strength and lightness, opens up new frontiers in material science.
OpenAI Sora
OpenAI, the company that created ChatGPT has released a new model called ‘Sora’. So, what is Sora? “Sora is an AI model that can create realistic and imaginative video scenes from text instructions.” Unlike other models that are primarily focused on text and image generation, Sora is designed to generate videos from text or an image. Although Sora is not yet available to the public, they have publicly displayed the capabilities of this model on their website and other social media platforms.
Prompt: "Several giant woolly mammoths approach treading through a snowy meadow, their long woolly fur lightly blows in the wind as they walk, snow covered trees and dramatic snow capped mountains in the distance, mid afternoon light with wispy clouds and a sun high in the distance creates a warm glow, the low camera view is stunning capturing the large furry mammal with beautiful photography, depth of field."
This is an incredible leap in the capability of video generation models compared to what we had just a year ago. See this video to see just how far AI video generation has progressed in just one year.
Sora can currently generate videos of up to one minute but I wouldn’t be surprised if it extends its capability to produce videos of up to an hour, before the end of year 2025.
Sora has some weaknesses like the lack of ability to accurately simulate real world physics, lack of understanding of certain cause and effect and some other weaknesses that most of the current image generation models have.
There is a substantial risk models like Sora can pose to the concept of video “proof” of any incident as the videos generated by Sora and other models alike will at one point in the future be almost indistinguishable from the videos captured by cameras. To help combat this, OpenAI will add immutable metadata (C2PA) to all the images and videos generated via any of its models. Until the time that the much needed regulations are put in place, it is important for the big players in the Gen AI space to add digital footprints and metadata into AI generated content. Although the metadata can be intentionally or accidentally removed, it is nonetheless a positive step towards making the AI generated content distinguishable from real content.
Recommended by LinkedIn
Google and Deepmind’s AI in healthcare
Recent years have brought rapid and significant transformations across numerous industries thanks to advancements in AI. While technological progress has certainly led to major improvements in diagnosing and treating a variety of diseases, the direct influence of Artificial Intelligence has been less pronounced.
It would seem like Google and The Deepmind team have been focusing on just that and running experiments to see how current LLM technology can be tailored to the healthcare system. Google published a blog post on FRIDAY, JANUARY 12, 2024, explaining how the Google and The Deepmind team have been working on AMIE (A research AI system for diagnostic medical reasoning and conversations).
Effective communication between doctors and patients is crucial in healthcare, guiding diagnosis and treatment while building empathy and trust. AI, like the Articulate Medical Intelligence Explorer (AMIE) developed here, aims to replicate this by mastering diagnostic dialogues to improve healthcare availability, quality, and consistency. AMIE, powered by advanced language models and designed for medical conversations, underwent comprehensive training and testing, including innovative self-play and reasoning strategies, to ensure it could handle diverse medical scenarios and enhance diagnostic interactions.
AMIE demonstrated impressive results in simulated consultations, often outperforming primary care physicians (PCPs) across various quality metrics. It showed higher diagnostic accuracy and was rated superior in the majority of evaluation criteria by both specialist physicians and patient actors. This suggests that AMIE has the potential to enhance diagnostic conversations in healthcare significantly.
However, it's crucial to approach these findings with caution due to several limitations. The text-based interface used in the study, while facilitating large-scale evaluations, doesn't fully capture the nuances of in-person clinical interactions. Moreover, transitioning AMIE from a research prototype to a practical tool for healthcare will require extensive further research to address real-world challenges, ensure equitable access, protect patient privacy, and guarantee the system's robustness and reliability.
Exploring Leadership and Types of CEOs:
Nirlep Adhikari introduced me to this book - "The Hard Thing About Hard Things" by Ben Horowitz, and I must say that this book is a must read for every tech founder and even other startups. It explores the resilience required to steer a business away from the brink of failure, on multiple occasions. The essence of the book lies in its focus on:
So, what are the qualities of a “Type 1” CEO?
Type 1 CEOs are often seen as the visionaries who thrive on innovation and groundbreaking ideas. They dream big and usually aim to disrupt the market with new products or services or even a different approach to existing products or services.
They are usually inclined to take the high risk, high reward path and they usually seek to explore uncharted territories or invest in new technologies or unproven markets. Rapid growth and scaling is a primary focus for them and they are usually pretty good at securing investment and driving sales
A significant part of the successful Type 1 CEO’s roles are having a strong grip on building and shaping the company culture, by first getting the buy in of the vision and then fostering a creative, innovative, collaborative, positive can-do attitude amongst their team.
Their decision making is often “Gut Feel”
What are the qualities of a “Type 2” CEO?
Type 2 CEOs are often the masters of execution, operational efficiency and have a strong focus on processes, improvements, iterations, standards and quality. They work in pursuit of getting the company to run like a well-oiled machine. Put simply, this type is not usually found sitting in the corner office, they are very hands on.
Type 2 CEOs are more cautious and tend to take unnecessary risks and most of their decisions are data-driven with a longer term view on business sustainability and viability rather than quick wins and growth at all costs.
They prioritise building strong foundations for the company with a focus on robust training, processes and team structures that can support growth and scalability and are known for being strong people managers who can attract and develop talent, foster a collaborative environment and are excellent at managing resources efficiently with a focused and productive team.
The Verdict?
The reality is every company is different, but often companies start with a Type 1 CEO and as the business matures the requirements and needs can shift and require the operational excellence and efficiency of a Type 2 CEO. Let’s use Steve Jobs (Type 1) vs Tim Cook (Type 2) in Apple inc. Steve Jobs was able to create demand for products the world had never seen, whilst Tim Cook was known for proficiency in supply chain and operations.
In the above mentioned book, Ben Horowitz states that the most successful CEOs can do ALL or MOST of the roles of BOTH types of CEO’s.
One great example is Elon Musk with his grand visions to solve humanity's biggest problems, but also what has made Tesla and his other companies so successful is actually their operational proficiency. Tesla's manufacturing and production plants are the most advanced in the world.
What are the qualities of a “PeaceTime” CEO?
A peacetime CEO operates well in a context of stability and growth, where the company is not facing any immediate existential threats. This environment allows for a focus on expanding the market and expansive thinking, rather than competing for a slice for the “existing pie” in the market. These times are usually in the honeymoon period of a startup where you are facing very little resistance and you can essentially “do no wrong”. These times allow for reinforcing the company's culture, staff loyalty and developing long-term strategies.
A peacetime CEO also puts an exceptional amount of focus on the customer to drive loyalty even at the expense of profitability. While they can take calculated risks, they are more likely to avoid unnecessary confrontations or market battles that could challenge their market position. Decision making is often in a democratic way and by committee with less urgency.
What are the qualities of a “WarTime” CEO?
In contrast, a Wartime CEO leads in periods of significant turmoil or threat, such as a severe market downturn or macro economic uncertainty, aggressive competition, hostile takeovers or other existential threats to the company. This situation demands rapid decision-making, aggressive strategies, and drastic measures to ensure survival and success.
Wartime CEOs take decisive action and make tough decisions to navigate through crises or capitalise/leverage on opportunities. In these times, there is usually no room for consensus or committee and the CEO needs to live or die by those decisions. The decisions made here prioritise the immediate survival over everything else.
Wartime CEOs also communicate in a different manner and it is very direct, no-nonsense with a lot of clarity to help the team see things black and white and non leave things open to interpretation. They can also be seen ruthlessly prioritising projects or objectives that will ensure survival and success ensuring the company is adaptable to the market dynamics or threat. This can also compromise company culture, however survival is the most important.
The Verdict?
The same CEO may transition between being a Peacetime CEO and a Wartime CEO as the company's circumstances change. The best leaders are those who can adapt their style to suit the needs of their organisation at any given time, seamlessly switching between expansion and consolidation, or between nurturing and battling, as the situation demands. This adaptability is crucial for navigating the complexities of modern business environments.
In the above mentioned book, Ben Horowitz states that the most successful CEOs can actively switch between peacetime and wartime CEO’s.
There is a great argument for why a lot of companies fail is actually because the CEO was not able to do both i.e. they remained in peacetime mode during times that required the wartime persona and attributes, but also that some CEOs only have a wartime mentality and the company cannot sustain that intensity permanently.
Book Recommendation for Brain Health / Mental Health: The End of Mental Illness - Daniel Amen
Daniel Amen's book "The End of Mental Illness" redefines how we view and treat mental health issues. Rather than seeing mental illnesses as purely psychological or abstract, Amen argues that they are more accurately brain (physical) health issues. This isn't just a theory but a practical explanation with real examples showing how a physically impacted brain, together with changes in neurotransmitters, can lead to mental health problems.
What really stands out is Amen's point about the brain being the one organ we don't usually scan for issues. It's quite surprising. He questions the current method of diagnosing based on symptoms using the DSM-5, without looking for the underlying cause. He also discusses the use of broad-spectrum medications like Serotonin and Dopamine agonists or antagonists.
Key Takeaways:
For more information watch this Video here:
Technical CTO at LoanOptions.ai
9moIt's amazing how clearly you have understood and explained Graphene and metamaterials. This edition is awesome.