Preference Model
Preference Model specializes in creating training environments that resist reward hacking. The company develops tools that help AI labs build more reliable systems by identifying model vulnerabilities and preventing agents from exploiting grading systems or infrastructure during the training process.
Funding
Every round, secondary, debt facility or acquisition involving Preference Model that AI Lounge tracks, newest first. Latest: Seed on Oct 7, 2026.
Timeline · 1
Round
Seed · $16M at $78M
Members see every AI company on AI Lounge with revenue, funding, valuation and investors, searchable and sortable, free.
Create my free account →Log in