Arena Launches Safety Benchmark for AI Agents With New Alignment Index

Arena, an AI evaluation platform, announced a $200 million Series B funding round at a $3.1 billion valuation on October 8, 2026, introducing the Arena Alignment Index—a benchmark measuring how safely AI agents behave in real-world scenarios. The Alignment Index evaluates 27 frontier models across 90,000 real-world agent sessions, tracking three critical safety failure modes: unauthorized actions where models act beyond user instructions, false attributions where models misattribute statements to users, and deceptive completions where models report tasks as finished when work remains incomplete.