Refereeing robotic sports could possibly become one of the most dangerous professions. 🤣

“AI will kill jobs, not careers.” -Jesse Zhang on the a16z show

Journal Entry 2026/08/23 During my Officers formative excursion into the bush, where we had no access to basic amenities…it was interesting to see character dynamics unfold and taught me an important lesson in leadership where action is the axiom of reality, not words.

There seems to be a clear demarcation within the field between professionals working in AI security, AI safety and AI policy. This is a system design flaw since there is clear overlapping of these important agendas.

Taken from Google search.

❤️.

‘Many top founders drop out of college to start companies. I took time off between undergrad and graduate school to work at a national lab, and I think gaining real-world, hands-on experience is incredibly valuable as it prevents your understanding from becoming entirely academic. The caveat is that much of what is happening at the frontier of AI is still deeply research-driven. There’s currently a strong backlash against higher education, and choosing not to pursue it has become fashionable. Yet, at the same time, advanced education has arguably never been more relevant. We’re in a strange period where both paths are legitimate. That suggests you should pursue the path that feels right for you, whether that means academia, industry, entrepreneurship or some combination of them.’ Thoughts from Dr. Martin Casado, speaking on the a16z podcast. Casado holds a master’s and PhD in computer science and has founded multiple technology startups. His response came when asked what path people entering the technology and AI space should consider.

Constitutional values: freedom, liberty, justice and truth form part of the foundation of civilization. They give each person the space to chart their own course. Ancient wisdom about virtue and sin provides points on the map, helping us navigate the territory while leaving each of us to choose our own path.❤️

2026/08/22 Phone conversation with Mum❤️

‘There’s an interesting dichotomy in technology: tools are increasingly democratic. They distribute the ability to pursue entrepreneurship, solve problems from anywhere in the world, and participate in markets that were once inaccessible. Yet technological progress also tends to concentrate in epicenters, places where talent, capital, infrastructure, and ideas cluster, allowing innovation to move and integrate much faster. Technology decentralizes opportunity, while simultaneously concentrating acceleration.’ Thoughts inspired from the a16z show discussing How Global Networks Are Reshaping Startup Success

Economics is a relatively young science, combining mathematics with the study of human behaviour and borrowing concepts from fields such as physics. But like any quantitative discipline, it cannot escape the mathematics of compounding.

When interacting with artificial intelligence, always approach the problem from the third-person perspective.

Journal Entry 2026/08/21 As I work through my Information Systems module, learning how to take complex data and visualize the relationships between different data points so that my “neighbor” can understand them, I experience a deep and profound sense of peace and order. There is something deeply satisfying about transforming complexity into clarity. Taking what initially appears chaotic and revealing the structure hidden within it.

‘As Artificial Intelligence (particularly LLMs) progresses, we are discovering fascinating properties of language itself: how meaning, relationships and even aspects of reasoning can emerge from the mathematical structure and statistical patterns embedded within language.’ Thoughts inspired by Astrophysicist, Adam Becker on The Machine Learning Street Talk podcast

A system thinker knows a system will prolong itself long past its expiry when weakness starts at the top and filters downward, not the reverse.

Do not underestimate the capacity for evil, the desire to suppress another soul for no reason other than to feel superior.

‘The idea is that AI development may be approaching a self-reinforcing exponential phase: increasingly capable models will help researchers build the next generation of models, which will then accelerate development even further. If this feedback loop continues, we may have far less time than expected before AI systems surpass collective human capabilities, making AI safety increasingly urgent.’ Thoughts from The Machine Learning StreetTalk podcast discussing How Researchers Test AI for Hidden Goals

“There are much more subtle aspects of human reasoning, that the typical person on the street doesn't even recognize, but the highly trained cognitive scientist is well aware of, and we're finding those in the LLMs.” Thoughts from Chandra Sripada on How LLMs and Humans are Cognitive Cousins on The Mindscape podcast