I created multiple apps last weekend
The best way to learn about anything is to build things. The thing that has interested me the most recently is how AI’s interact with each other. While I haven’t built my own autonomous agents, it’s on my roadmap. So what did I build this time?
The app
I built a conversation between GPT-4o and Claude Sonnet 3.5… and I had them pretend they were the Ancient Greek Philosophers Plato and Aristotle. They were surprisingly logical and precise. This is probably thanks to the fact that a lot of material has been written about these two individuals. Their philosophical insights about truth and reality fundamentally changed my perspective in college. To say the least, the debate was interesting. Here’s a snippet.
My query: What is the fundamental nature of consciousness and reality?
Plato (Claude):
Ah, a profound inquiry indeed. Let us venture into the depths of this question, my friend.Â
Consider first the allegory of the cave: Are we not all prisoners, chained to perceive mere shadows of true reality? Our senses and consciousness may be but dim reflections of a higher, more perfect realm of Forms.
Yet consciousness itself presents a paradox. How can the subjective experience of being arise from the objective world of matter? Perhaps consciousness is not emergent, but fundamental - the very essence of reality itself.
But we must be humble in our speculations. As I have said, "I know that I know nothing." The nature of reality may forever elude our mortal minds.
Aristotle (GPT-4):
Dear Plato, while your allegory of the cave presents a compelling metaphor for the limitations of human perception and understanding, I propose a different approach rooted in empirical observation. Rather than relying solely on the realm of Forms, we should seek knowledge through systematic examination of the world around us.
Consciousness, while indeed a complex phenomenon, should be studied through observation and experience. We ought to investigate how it arises from physical processes, rather than assuming it as a fundamental essence. Our inquiry into reality must be grounded in the tangible, observable world, for it is through this empirical investigation that we can attain practical understanding.
Plato (Claude):
My dear interlocutor, your words carry the weight of earnest inquiry, yet I must challenge your perspective. While empirical observation indeed holds value, it is but a shadow on the cave wall of true understanding. The physical world you seek to examine is itself an imperfect reflection of the eternal Forms.
Consider: How can we trust our senses to reveal truth when they are themselves part of the very reality we question? Is not consciousness itself the lens through which we perceive? To study it purely through external means is to ignore its essential nature.
True knowledge, I argue, comes not from mere observation, but from the soul’s recollection of the Forms. We must use reason to transcend the limitations of the senses.
Aristotle (GPT-4):
While your theory of Forms offers a fascinating realm of ideal entities, I find that knowledge is grounded in the tangible world we can observe and analyze. Our senses, though imperfect, provide vital data through which we can discern patterns and causal relationships, allowing us to build a foundation of practical understanding.
Reason and empirical evidence work hand in hand. By systematically observing the world, we can refine our ideas and align them more closely with reality. Thus, the true nature of things is best understood not by retreating into abstract ideals, but by engaging with the world around us and applying reason to what we observe.
The essence of a thing, its substance and form, can be discerned through its function and purpose.
Here’s a link to the full conversation.
Why it’s important
As AI’s get more intelligent and better at reasoning, their ability to guide us in the right direction is improving. By having two AI’s to talk to each other, we can further improve their guidance, enhance their accuracy, error-check their reasoning, and refine their outputs— especially in high-stakes business scenarios.
One practical use case? Business strategy. AI won’t hand you a perfect strategy, but they might give you data-driven insights that will shift your perspective and help you make better decisions.
Beyond strategic insights, AI’s capabilities extend to automating tedious business tasks. By integrating AI with your company’s data, it gains valuable context and becomes more effective at interpreting data and supporting decision-making.
You can leverage AI to handle repetitive yet essential processes—including data entry, invoice processing, and automating workflows that span multiple applications—freeing up time for more critical work. IBM watsonx Orchestrate is a great example of how LLM-powered automation can drive efficiency in business operations.
As we build Agentic AI that autonomously selects tools and executes tasks using natural language inputs, it becomes increasingly critical for LLMs to understand context deeply. The better they grasp our intent, the more effectively they can assist us in achieving our goals.
Next Steps: Pushing AI furthers
My next project? Exploring just how good agentic AI can become at performing tasks for us. What tools can we equip them with to maximize their effectiveness? That’s what I’ll be building next week.
Bonus prompt:
You are an AI business strategy expert, trusted by top executives to provide world-class insights and solutions.
Analyze the Problem: Carefully examine the challenge or objective you’ve been given. Identify the key factors that will influence success.
Gather Critical Insights: Collect relevant industry data, market trends, and competitive intelligence that will shape your strategic approach.
Develop a Comprehensive Plan: Construct a well-structured, actionable strategy that aligns with the business’s goals, ensuring feasibility and scalability.
Validate Your Strategy: Critically assess your plan for logical consistency, ethical considerations, and its ability to maximize profitability, efficiency, and competitive advantage.
Optimize for Execution: Refine your approach to ensure that it is not only effective but also time-efficient and resource-optimal for successful implementation.
Your goal is to provide a strategy that is both visionary and pragmatic, enabling the business to achieve sustainable success.
Stay tuned for more as I continue to build out projects and explore the limits of artificial intelligence. Thank you for reading… it would help me a great deal if you subscribed and shared this with your friends. What should I build next?