In Search of the Ingredients of Open-Endedness: Replicating Picbreeder with Large Vision-Language Models
This paper explores replicating Picbreeder using large Vision-Language Models to understand open-ended evolution. The research delves into the essential components needed for systems to continuously generate novel and complex outputs without explicit goals. This work was published on arXiv by Sam Earle and five co-authors.
A new paper, "In Search of the Ingredients of Open-Endedness: Replicating Picbreeder with Large Vision-Language Models," explores the exciting field of open-ended evolution in artificial intelligence. This research aims to understand how AI systems can continuously generate novel and complex outputs without explicit, predefined goals. The study specifically investigates the replication of Picbreeder, a platform known for its open-ended evolutionary artwork, using advanced Large Vision-Language Models (LVLMs). This approach could shed light on the fundamental mechanisms driving creativity and unpredictable innovation in AI. The paper was authored by Sam Earle and five collaborators and was published on arXiv on April 1, 2026. The research highlights the ongoing efforts within the scientific community to push the boundaries of AI capabilities beyond narrow task-oriented applications. The findings could have significant implications for the development of more autonomous and creative AI systems in the future.
Related articles
The AI world is getting ‘loopy’
AI models are taking a significant leap forward with the adoption of "agentic loops," where AI agents continuously prompt each other to improve code and solve complex problems. This approach, though potentially resource-intensive, promises to unlock new levels of autonomous problem-solving and efficiency in AI applications.
Codex-maxxing for long-running work
Codex is increasingly being used by organizations to support long-running projects that go beyond a single prompt. This whitepaper by Jason Liu offers practical strategies for leveraging Codex as a persistent workspace, managing complex workflows and sustaining progress.
Nobel laureate John Jumper is leaving DeepMind for rival Anthropic
Nobel laureate John Jumper is departing Google DeepMind to join its competitor, Anthropic, after dedicating nearly nine years to DeepMind, where he led the AlphaFold team. Jumper, who shared a Nobel Prize for his work on AlphaFold, expressed gratitude for his time at DeepMind while looking forward to new endeavors.