Standard AI benchmarks are an engineering tool for measuring performance. A more scientific approach, borrowed from cognitive psychology, uses targeted experiments. By designing problems where specific patterns of success and failure are diagnostic, researchers can uncover the underlying mechanisms and principles of an AI system, yielding deeper insights than a simple score.

Traditional AI benchmarks fail to capture the value of models that enable entirely new capabilities. The concept of an 'unlock index' suggests we should evaluate models based on the new applications they make possible—like the visual proactivity of TML's interaction model—rather than just performance on existing tasks.

OpenAI's evals team is looking beyond current benchmarks that test self-contained, hour-long tasks. They are calling for new evaluations that measure performance on problems that would take top engineers weeks or months to solve, such as creating entire products end-to-end. This signals a major increase in the complexity and ambition expected from future AI benchmarks.

AI struggles with long-horizon tasks not just due to technical limits, but because we lack good ways to measure performance. Once effective evaluations (evals) for these capabilities exist, researchers can rapidly optimize models against them, accelerating progress significantly.

As benchmarks become standard, AI labs optimize models to excel at them, leading to score inflation without necessarily improving generalized intelligence. The solution isn't a single perfect test, but continuously creating new evals that measure capabilities relevant to real-world user needs.

Just as standardized tests fail to capture a student's full potential, AI benchmarks often don't reflect real-world performance. The true value comes from the 'last mile' ingenuity of productization and workflow integration, not just raw model scores, which can be misleading.

Traditional AI benchmarks are seen as increasingly incremental and less interesting. The new frontier for evaluating a model's true capability lies in applied, complex tasks that mimic real-world interaction, such as building in Minecraft (MC Bench) or managing a simulated business (VendingBench), which are more revealing of raw intelligence.

The rapid improvement of AI models is maxing out industry-standard benchmarks for tasks like software engineering. To truly understand AI's impact and capability, companies must develop their own evaluation systems tailored to their specific workflows, rather than waiting for external studies.

The true measure of a new AI model's power isn't just improved benchmarks, but a qualitative shift in fluency that makes using previous versions feel "painful." This experiential gap, where the old model suddenly feels worse at everything, is the real indicator of a breakthrough.