In April 2025, Tokyo-based artificial intelligence company Sakana AI announced a significant milestone: their AI system, "The AI Scientist-v2," autonomously authored a scientific paper that successfully passed peer review at the "I Can't Believe It's Not Better" workshop during the International Conference on Learning Representations (ICLR) 2025. This event marks the first instance of a fully AI-generated paper being accepted through the peer-review process.

Sakana AI, founded in July 2023 by David Ha, Llion Jones, and Ren Ito, focuses on AI research in evolution and collective intelligence. The company's name, "Sakana," meaning "fish" in Japanese, symbolizes the concept of simple entities forming a coherent whole, akin to a school of fish. Llion Jones is notably one of the co-authors of the seminal paper "Attention Is All You Need," which introduced the Transformer model, a cornerstone in modern AI development.

The AI Scientist-v2 represents a significant advancement over its predecessor by eliminating reliance on human-authored code templates and generalizing effectively across diverse machine learning domains. It employs a novel progressive agentic tree-search methodology managed by a dedicated experiment manager agent. Additionally, the system integrates a Vision-Language Model (VLM) feedback loop to iteratively refine the content and aesthetics of figures within the manuscripts.

To evaluate The AI Scientist-v2, Sakana AI submitted three fully autonomous manuscripts to the "I Can't Believe It's Not Better" (ICBINB) workshop at ICLR 2025. Notably, one manuscript achieved reviewer scores sufficient to surpass the average human acceptance threshold, marking the first instance of a fully AI-generated paper successfully navigating peer review. This accomplishment underscores the growing capability of AI in conducting all aspects of scientific research.

Sakana AI conducted this experiment with the full cooperation of both the ICLR leadership and the organizers of the ICBINB workshop. The company emphasized transparency and ethical conduct throughout the process. In a statement, Sakana AI noted, "We conducted this experiment with the full cooperation of both the ICLR leadership and the organizers of an ICLR workshop."

The AI-generated paper's acceptance has sparked diverse reactions within the scientific community. Some experts view it as a significant step toward Artificial Research Intelligence (ARI), where machines autonomously conduct research. However, others express concerns about the potential narrowing of research scope and the loss of diverse perspectives that fuel innovation. Anthropologist Lisa Messeri and psychologist M.J. Crockett cautioned against treating "AI products as autonomous researchers," emphasizing that "producing papers and producing knowledge are not the same."

The success of The AI Scientist-v2 highlights the potential for AI to significantly accelerate scientific discovery. By automating various aspects of research, AI systems could enhance productivity and enable unprecedented scalability in knowledge generation. However, this development also raises questions about the role of human scientists, the integrity of the peer-review process, and the ethical considerations of AI-generated research.

The acceptance of an AI-generated paper at a reputable workshop signifies a pivotal moment in the integration of artificial intelligence into scientific research. While it showcases the remarkable capabilities of AI, it also prompts a broader discussion on the future of scientific inquiry, the preservation of ethical standards, and the balance between human and machine contributions in the pursuit of knowledge.