About Us

In 2000, Thomas Kören, MS, ScD and Peter Gray, MS, ScD, both graduates from MIT's Computational Data Optimization (CDO) master's program and Computational Science and Engineering (CSE) doctoral program, embarked on a collaborative research project focused on deterministic statistical models enhanced by artificial intelligence algorithms. To introduce an element of randomness within their Cardano probability space, they frequently incorporated lottery draws into their mathematical models. After two years of rigorous research and observation, their efforts culminated in a groundbreaking discovery: the apparent non-existence of true randomness.

Intrigued by the persistent patterns in their models, Kören and Gray traced the source to the lottery data itself. Initially suspecting data contamination or manipulation, they sourced lottery data from diverse global lotteries, only to find consistent patterns each and every time. This unexpected consistency led them to apply their advanced AI algorithms to isolate these patterns within the lottery data. Despite their AI's sophistication, it lacked the necessary power to fully uncover this phenomenon.

Undeterred, Kören and Gray dedicated themselves to developing a cutting-edge AI neural network capable of identifying, replicating, and predicting these patterns. After another two years of intense development, they succeeded in 2004 with the creation of the NeuralAnalyzer i7000 lottery system. This sophisticated system, reserved for their research purposes, demonstrated that seemingly random processes like lotteries possess a detectable "memory".

By 2006, facing dwindling research funds, Thomas and Peter accepted positions at a multinational firm in Mexico. Relocating to Mexico City, they continued their research part-time while fulfilling their new professional roles. In 2008, after extensive upgrades and refinements to the NeuralAnalyzer i7000, their efforts paid off spectacularly. In the summer of 2009, their system secured a jackpot in Mexico's Melate 6/56 lottery, earning them over US$3.7 million. Their success persisted, and by April 2012, their winnings exceeded US$4.9 million from over 540 lottery hits.

With a renewed research budget, Dr. Thomas Kören and Dr. Peter Gray set out to definitively prove that inherent patterns exist in every lottery, negating the concept of pure randomness. They spent the next four years perfecting their AI neural network model. Their diligence culminated in an unprecedented breakthrough: the first generation NeuralReality AI Engine, a neural network that boasted 256 cores, 4 NVIDIA Tesla K40 GPUs, and 21.87 TeraFLOPs of processing power. This advanced system, which featured over 1 million active neurons and data from 170 global lotteries, enabled Thomas and Peter to earn over US$38.6 million in a single month.

This remarkable achievement led to the launch of neural-lotto.net.

Today, Neural-Lotto stands as a monument to three unique and unparalleled AI neural networks designed for advanced pattern analysis, offering unprecedented lottery predicting capabilities of unrivalled precision and accuracy to the general public for the first time. These technologies differ considerably in architecture, infrastructure, and operating principles, ranging from classical to quantum computing.

The first of them is the upgraded and significantly enhanced NeuralReality³ AI Engine (standard) processing infrastructure. It operates on a high-performance 64-bit ARMv8.2-A Scalable Vector Extension SIMD instruction set, featuring a 512-bit vector implementation architecture with Tensor Core GPUs. It includes 512 P/SDM, 128 search and discovery modes, 25 algorithms, and 24+ million neurons. Compared to 2018's KGA6, the current NeuralReality³ AI Engine is up to 3.4 times more powerful (217.62 vs. 63.74 TFLOPs). This system is ideal for data scientists, AI researchers, and lottery system enthusiasts, offering extensive (hyper)parameter user controls to cater to their technical requirements.

In contrast, the KGA7 Phaedra DAU processing system utilizes 4 nodes based on Fujitsu's DAU (Digital Annealing Unit) architecture, which features the world's first quantum-inspired digital technology. Each node comprises 4 DAU CPUs and 4+2 IBM Power10 hemispheric control CPUs, interconnected via PowerAXON. Unlike the NeuralReality³ AI Engine, the KGA7 Phaedra DAU requires minimal user adjustments, as it features 4 genetic self-evolving, self-adaptive hypercube-based geometric 4D deep learning AI reasoning units, employing both inductive and deductive methods.

Lastly, the KGA8 Phaethon Quantum processing system leverages IBM's Q System One 100+ qubit utility-scale quantum computer, soon to be joined by System Two, operating at full system capacity using the IBM Quantum Accelerator package. It also runs parallelly on the D-Wave 5th generation Advantage 5,000 qubit quantum computer. The results from both quantum systems are analyzed and combined to deliver the highest possible lottery prediction accuracy. The KGA8 Phaethon Quantum system is designed to require no user (hyper)parameter or variable settings, making it accessible and user-friendly for individuals without technical expertise.

At their core, all three Neural-Lotto lottery predictors are intelligent learning systems, embodying the fundamental principles of all neural networks. However, they stand unparalleled as the most advanced search, discovery, and pattern-seeking artificial intelligence deep multi-neural networks ever applied to lotteries. No other lottery prediction system in existence can match their capabilities. The more real-world lottery data and specific inputs these AI systems process, the more they enhance their learning, efficiency, and performance, setting benchmarks that are unmatched at any price.

Well then, what in the world are you waiting for to put these monumental systems to work for you?