Valve uses a machine learning model to assign every Counter-Strike player an invisible "trust score" based on their behavior across their entire Steam account history. Players with low scores are quietly matched against each other rather than being banned outright, which limits the impact of cheaters and toxic players on the wider community. The system has been running since November 2017.

Trust Factor draws on signals from across a player's Steam account — frequency of cheat reports, presence of linked accounts with prior bans, total hours played, activity in other games, and other undisclosed inputs — and feeds them into a machine learning model to predict the likelihood of disruptive behavior. Valve deployed the system six months before announcing it publicly. A 2018 patent application (US20200078688A1) and international filing (WO2020051517A1), both titled "Machine-Learned Trust Scoring for Player Matchmaking," describe the technical approach in detail and explicitly state that trained machine learning models outperform rules-based systems for predicting behaviors like cheating, match abandonment, and abusive language. At GDC 2018, Valve engineer John McDonald showed that players complaining about cheaters in their matches were consistently found to have low trust scores of their own, many linked to dozens of banned accounts.