Blatant AI slop wins $25k DeepMind Kaggle AGI competition

The Kaggle Measuring AGI competition awarded the $25,000 grand prize, sponsored by DeepMind, to a model named Blatant AI slop — the first prize announced on the competition’s discussion board [Kaggle Forums].

The contest asked participants to build models capable of quantifying progress toward Artificial General Intelligence, a long‑standing goal in the field. Judges evaluated submissions on a set of benchmark tasks designed to probe generalization across domains.

Blatant AI slop secured the top spot after demonstrating consistent performance across the benchmark suite, meeting the judges’ criteria for robustness and scalability [Kaggle Forums]. The model’s name, chosen by its creators, belies the technical rigor that earned it the prize.

The announcement sparked a flurry of commentary on the Kaggle forum, where competitors dissected the model’s architecture, training regime, and evaluation metrics. Threads quickly filled with technical critiques, suggestions for improvement, and broader debates about how such competitions shape AGI research.

The win underscores three concrete developments: first, the strong engagement of the AI community in a high‑stakes, open‑source challenge; second, the emergence of novel measurement techniques that could inform future AGI benchmarks; and third, DeepMind’s continued investment in community‑driven research, signaling corporate confidence in the competitive model as a catalyst for progress. These factors together illustrate how public contests can accelerate methodological innovation while drawing industry support into the open‑research ecosystem.