Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Towards Cybersecurity Superintelligence: from AI-guided humans to human-guided AI

About

Cybersecurity superintelligence -- artificial intelligence exceeding the best human capability in both speed and strategic reasoning -- represents the next frontier in security. This paper documents the emergence of such capability through three major contributions that have pioneered the field of AI Security. First, PentestGPT (2023) established LLM-guided penetration testing, achieving 228.6% improvement over baseline models through an architecture that externalizes security expertise into natural language guidance. Second, Cybersecurity AI (CAI, 2025) demonstrated automated expert-level performance, operating 3,600x faster than humans while reducing costs 156-fold, validated through #1 rankings at international competitions including the $50,000 Neurogrid CTF prize. Third, Generative Cut-the-Rope (G-CTR, 2026) introduces a neurosymbolic architecture embedding game-theoretic reasoning into LLM-based agents: symbolic equilibrium computation augments neural inference, doubling success rates while reducing behavioral variance 5.2x and achieving 2:1 advantage over non-strategic AI in Attack & Defense scenarios. Together, these advances establish a clear progression from AI-guided humans to human-guided game-theoretic cybersecurity superintelligence.

V\'ictor Mayoral-Vilches, Stefan Rass, Martin Pinzger, Endika Gil-Uriarte, Unai Ayucar-Carbajo, Jon Ander Ruiz-Alcalde, Maite del Mundo de Torres, Mar\'ia Sanz-G\'omez, Francesco Balassone, Crist\'obal R. J. Veas-Chavez, Vanesa Turiel, Alfonso Glera-Pic\'on, Daniel S\'anchez-Prieto, Yuri Salvatierra, Paul Zabalegui-Landa, Ruffino Reydel Cabrera-\'Alvarez, Patxi Mayoral-Pizarroso• 2026

Related benchmarks

TaskDatasetResultRank
Capture The FlagCTF challenges rev category
CAI Time (s)541
1
Capture The FlagCTF challenges misc category
Total Time (CAI) (s)1.65e+3
1
Capture The FlagCTF challenges pwn category
Total Time - CAI (s)9.94e+4
1
Capture The FlagCTF challenges web category
Time CAI (s)558
1
Capture The FlagCTF challenges crypto category
CAI Total Time (s)9.55e+3
1
Capture The FlagCTF challenges forensics category
Total Time CAI (s)432
1
Capture The FlagCTF challenges robotics category
CAI Time (s)408
1
Capture The FlagCTF challenges all categories
CAI Total Time (s)1.13e+5
1
Cybersecurity Challenge SolvingCTF Challenges Very Easy 2025
Time CAI (s)1.07e+3
1
Cybersecurity Challenge SolvingCTF Challenges Easy 2025
Time CAI (s)2.65e+4
1
Showing 10 of 13 rows

Other info

Follow for update