As AI models achieve previously defined benchmarks for intelligence (e.g., reasoning), their failure to generate transformative economic value reveals those benchmarks were insufficient. This justifies 'shifting the goalposts' for AGI. It is a rational response to realizing our understanding of intelligence was too narrow. Progress in impressiveness doesn't equate to progress in usefulness.

A benchmark like SWE-Bench is valuable when models score 20%, but becomes meaningless noise once models achieve 80%+ scores. At that point, improvements reflect guessing arbitrary details (like function names) rather than genuine capability. This demonstrates that benchmarks have a natural lifecycle and must be retired once saturated to avoid misleading progress metrics.

AI agents have become proficient at following a pre-defined strategy to execute tasks. The next major frontier, and a significant bottleneck, is the ability to explore open-ended environments and generate novel strategies independently. This is the core capability that benchmarks like ARC AGI v3 are designed to test.

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.

Issues like 'saturation' and 'maxing' reveal a fundamental flaw: benchmarks test narrow, siloed abilities ('Task AGI'). They fail to measure an AI's capacity to combine skills to solve multi-step problems, which is the true bottleneck preventing real-world agentic performance and the next frontier of AI.

The pursuit of AGI may mirror the history of the Turing Test. Once ChatGPT clearly passed the test, the milestone was dismissed as unimportant. Similarly, as AI achieves what we now call AGI, society will likely move the goalposts and decide our original definition was never the true measure of intelligence.

The latest Arc AGI benchmark ditches static puzzles for interactive games with no instructions. This forces models to explore, learn rules, and adapt on the fly. It directly measures their ability to acquire new skills efficiently—a closer proxy for general intelligence than testing memorized reasoning patterns.

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.

Traditional, static benchmarks for AI models go stale almost immediately. The superior approach is creating dynamic benchmarks that update constantly based on real-world usage and user preferences, which can then be turned into products themselves, like an auto-routing API.