AI Conferences Struggle With Surge of Poor-Quality Submissions
Artificial intelligence (AI) researchers and conference organizers are grappling with a growing problem: a deluge of low-quality, AI-generated submissions. The influx of what some in the field have dubbed “slop” — hastily generated content lacking originality or scientific rigor — is overwhelming peer review systems and prompting significant changes to submission guidelines at major conferences.
Several leading AI conferences, including the International Conference on Learning Representations (ICLR) and the Conference on Neural Information Processing Systems (NeurIPS), have introduced strict new rules to curb the spread of AI-generated content. These measures are a response to the rapid rise in the use of large language models (LLMs) such as ChatGPT to produce entire papers or peer reviews with minimal human oversight.
Conference Organizers Impose Restrictions on AI Tools
Organizers of top-tier AI gatherings have started to mandate that authors disclose any use of generative AI tools in the creation of their submissions. Some conferences now explicitly prohibit the use of LLMs to write or review papers, citing concerns over accuracy, bias, and the erosion of academic integrity.
“We’re seeing a dramatic increase in the number of submissions that appear to be written by AI,” said one conference organizer. “Many of these lack the depth, novelty, or clarity expected in academic research.”
The changes reflect a growing unease across the AI research community about the unintended consequences of their own technologies. While LLMs have made it easier than ever to draft coherent text, they often produce content that is superficially plausible but substantively flawed. This has led to a surge in submissions that waste reviewers’ time and dilute the quality of conference proceedings.
Peer Review Systems Under Strain
The peer review process, which relies on volunteer researchers to evaluate submissions, is buckling under the weight of the increased volume. Reviewers report spending more time sifting through AI-generated manuscripts that lack meaningful contributions or demonstrate fundamental misunderstandings of the field.
“It’s not just that these papers are bad,” said a reviewer for a major AI conference. “It’s that they’re bad in the same ways — generic language, unsupported claims, and a lack of real insight. It’s clear they’re being produced by machines.”
Some reviewers have also raised concerns about AI-generated peer reviews, which may be used by overwhelmed researchers who are assigned dozens of papers to evaluate in a short time. These reviews can be vague, inaccurate, or misleading, undermining the credibility of the process.
Balancing Innovation and Integrity
While many in the AI community recognize the potential of generative tools to assist with tasks like editing and summarization, most agree that they should not replace human authorship or critical thinking. The challenge lies in finding the right balance between leveraging AI for efficiency and preserving the standards of academic scholarship.
“There’s a role for AI in the research process,” said an AI ethics expert. “But we need to be transparent about when and how it’s used. Otherwise, we risk eroding trust in the scientific enterprise.”
To that end, conferences are experimenting with new submission protocols. Some are using AI detectors to flag suspicious submissions, while others are requiring authors to submit statements detailing their use of generative tools. There is also talk of incorporating AI literacy into reviewer training, to help evaluators better identify machine-written text.
Community Response and Future Outlook
The response from the research community has been mixed. Some applaud the new measures as necessary steps to protect the integrity of the field. Others worry that overly strict rules could stifle innovation or penalize non-native English speakers who rely on tools like ChatGPT to improve their writing.
“We don’t want to throw the baby out with the bathwater,” said a senior AI researcher. “These tools can help democratize access to research and reduce barriers to participation. But we need to use them responsibly.”
As AI-generated content becomes more sophisticated, the line between human and machine authorship will only blur further. Experts say that ongoing vigilance, clear guidelines, and a commitment to ethical standards will be essential to navigating this evolving landscape.
In the meantime, conference organizers are urging researchers to uphold the values of originality, transparency, and scientific rigor — whether their papers are written by humans, machines, or a combination of both.
This article is inspired by content from Original Source. It has been rephrased for originality. Images are credited to the original source.