AIEvaluating AI in games needs a trust layer, not another leaderboardEvaluating AI in games has become one of the harder problems in applied AI, and the… Geoff RAugust 3, 2026August 3, 2026
Ontology EVMOntology MainNet v3.1.2: Four Ethereum Opcodes Arrive on the Ontology EVMThe next Ontology MainNet upgrade, to v3.1.2, is scheduled for block height 20,800,000, estimated to… Geoff RJuly 29, 2026July 30, 2026
DataAIOntology at Eight: Building the Verified Human Data Layer for the AI EconomyEight years ago, on 30 June 2018, the Ontology MainNet went live. It has run… Geoff RJuly 3, 2026July 3, 2026
Evaluating AI in games needs a trust layer, not another leaderboard AI Geoff RAugust 3, 2026August 3, 202614 mins0 Evaluating AI in games has become one of the harder problems in applied AI, and the industry’s current…continue reading..
Ontology MainNet v3.1.2: Four Ethereum Opcodes Arrive on the Ontology EVM Ontology EVM Geoff RJuly 29, 2026July 30, 202610 mins0 The next Ontology MainNet upgrade, to v3.1.2, is scheduled for block height 20,800,000, estimated to be on…continue reading..
Ontology at Eight: Building the Verified Human Data Layer for the AI Economy Data Geoff RJuly 3, 2026July 3, 202610 mins0 Eight years ago, on 30 June 2018, the Ontology MainNet went live. It has run without interruption…continue reading..
When human oversight becomes a compliance requirement AI Geoff RJune 22, 2026June 22, 202612 mins0 Human oversight documentation is the auditable record proving that real, identifiable people reviewed and shaped a model during…continue reading..
When the judge shares the blind spot Data Geoff RJune 15, 2026June 15, 202615 mins0 LLM-as-judge blind spots are the systematic reasoning failures an automated evaluator inherits from the model it is…continue reading..
Evaluator-backed benchmarking: your benchmark is only as good as your evaluators Data Geoff RJune 9, 2026June 9, 202613 mins0 Evaluator-backed benchmarking is the structural counter to benchmark gaming. When the underlying evaluators carry verifiable identity, longitudinal…continue reading..
Reward models need reward-model QA Data Geoff RJune 8, 2026June 8, 202612 mins0 Reward model QA is the missing layer that turns step-level preference data into trustable training signal. When…continue reading..
The evaluator uniqueness primitive: from sybil resistance to agent evaluation Data Geoff RJune 5, 2026June 5, 202616 mins0 Evaluator uniqueness, the property that one person can prove they are one unique evaluator without disclosing identity,…continue reading..
Continuous training needs continuous evaluators Data Geoff RJune 4, 2026June 4, 202614 mins0 Longitudinal evaluation is the human-judgement layer that scales alongside continual model adaptation. A continually retrained model paired…continue reading..
Your reward model is only as good as your preference data Data Geoff RJune 3, 2026June 3, 202613 mins0 Preference data integrity is the upstream gate that determines what every distilled, fine-tuned, or RLHF-aligned model is…continue reading..
