AI · 2h ago
AI Research Engine Discovers Power Law in Twin Prime Residuals
A developer built an autonomous AI research engine called Luka and applied it to the twin prime conjecture. Luka found that the residual error in the Hardy-Littlewood formula follows a perfect power law with R² = 0.9997. The AI also falsified a recent oscillatory model, showing predictions off by factors of 100–700.
Meridian48 take
While the mathematical discovery is intriguing, the real story is Luka's ability to autonomously test hypotheses and falsify models, hinting at a future where AI aids mathematical research.
Read the full reporting
I Built an AI Research Engine and It Found a Perfect Power Law in Twin Primes →
DEV Community
ai-researchmathematics