Singapore Inventor Unveils Pieces of Eight (P8): AI That Learns from Just Eight Samples

From diagnosing rare diseases with only a few known cases to predicting economic shifts, P8 makes possible what big AI can't.

SINGAPORE, September 17, 2025 /EINPresswire.com/ -- A groundbreaking new artificial intelligence model, Pieces of Eight (P8), has been unveiled by Singapore serial entrepreneur and inventor Kannappan Chettiar. Unlike conventional AI, which requires massive datasets and GPU power, P8 can achieve 90–95% accuracy with only about eight reliable samples. This shift from big data to small data AI signals a new era of efficiency, safety, and accessibility in intelligent systems.

Conventional AI systems, including Large Language Models or LLMs, typically require millions or billions of examples and massive computing power to reach real-world accuracy of 70–90%.

At the core of P8 is Binomial Relational Logic, a novel patent-pending framework that redefines learning as a process of mapping relationships among the known (X), the unknown (Y), and the stabilizing relational field (Z). Predictions are generated not through probability, but through resonant mapping, which aligns data with outcomes in a stable and safe manner.

This approach mirrors how humans learn naturally from limited experiences, rather than depending on brute force repetition and simulations. The innovation is embodied in Reflex AI Chips developed by Chettiar’s Indian company Fizix Solar Innovations, which execute P8’s logic directly in hardware for energy, solar, and power delivery applications.

What makes this achievement remarkable is not only the technology but also the story behind it. Chettiar is not an engineer by training. His formal background is in law, finance, and international arbitration, with a long career in education and corporate leadership. Only after retiring six years ago did he begin to teach himself engineering and computing from scratch — a journey driven by the vision that intelligence should be relational, not probabilistic. P8 is the result of this late-career reinvention: an inventor without a conventional technical background developing a new paradigm of AI and engineering.

“Today’s AI depends on overwhelming scale,” Chettiar said. “With P8, we show that intelligence can be learned in fewer than eight steps and at any age — as long as you build a relationship with the subject. In my case, that was engineering, computing, and AI.”

With immediate applications in renewable energy, healthcare, mobility, and consumer electronics, P8 and Reflex AI Chips promise safer, more efficient, and more democratic access to AI technology — made possible by an inventor who embodies lifelong learning and reinvention.

P8 makes possible what big AI cannot. From diagnosing rare diseases with only a few known cases to predicting economic shifts or conflict risks from limited signals, P8 thrives where data is scarce. It proves that intelligence does not need scale — only reliable relationships.

“Did you take a thousand routes to remember your way home, or just a few?” Chettiar asks. “Learning, even for machines, is not about abstraction but about relationships. P8 mirrors the way humans and nature learn — from very little, with great certainty.”

The announcement coincides with the release of Chettiar’s latest paper, which details the theory and applications of Small Data Predictive Models and their embodiment in hardware.

Kannappan K Chettiar
Switching Battery Inc.
+1 831-643-5919
email us here
Visit us on social media:
LinkedIn

Artificial Intelligence (AI) vs Quantum Intelligence (QI): Why Flow Matters | Kannappan Chettiar

Legal Disclaimer:

EIN Presswire provides this news content "as is" without warranty of any kind. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the author above.