RC RANDOM CHAOS

Decision Models Reshape SOC Triage, But Forensic Context Still Rules

· via The Hacker News

Original source

AI Decision Models and Their Impact on AI SOC Platforms

The Hacker News →

A new category of AI model called decision models went from nonexistent to widely available in about a month. TypeSafe AI released Jev on September 15, OpenAI opened a public beta for its Decisions API on October 6, and Cribl published an open LoRA adapter called cribl-decision. Unlike text-generating models, these take a state and typed questions and return a choice, score, or yes/no with a calibrated probability. For SOC alert triage, that means you can set thresholds—auto-close above 0.97, escalate below 0.6, and send the middle to an analyst. Jev costs $42 per billion input tokens with 70-to-500-millisecond responses, making per-alert analysis economically viable even at 450,000 alerts a year. Every verdict is a fixed-schema output that can be logged, replayed, and audited far more easily than a paragraph of generated reasoning.

Read the full article

Continue reading at The Hacker News →

This is an AI-generated summary. Read the original for the full story.