Author Benjamin Skuse partnered with the journal IEEE Access to run a war-game experiment testing how easily artificial intelligence can fabricate an entire research paper from scratch. Relying on tools like ChatGPT and Gemini, Skuse generated hypotheses, code, and text without performing any actual laboratory experiments or manual copyediting. The study found that prompting tools section-by-section produced far more detailed scientific content than generalized initial attempts. IEEE research-integrity manager Amanda Sulicz warned that such automation presents an escalating threat to the integrity of the broader scientific enterprise.
Key points
Free, publicly accessible artificial intelligence tools can successfully fabricate compre…: Free, publicly accessible artificial intelligence tools can successfully fabricate comprehensive scientific papers based on research that was never performed.
Standard academic publishing safeguards are struggling to keep pace with the massive infl…: Standard academic publishing safeguards are struggling to keep pace with the massive influx of automated, low-quality manuscripts.
Section-by-section prompting enables artificial intelligence models to bypass thin initia…: Section-by-section prompting enables artificial intelligence models to bypass thin initial outputs and generate sophisticated academic text.
Sources checked by our analysis pipeline: 1
I Used AI to Fake a Scientific Paper. Here’s What Happened
My experiment with an IEEE journal revealed AI’s threats to scholarly publishing
Science has a spam problem. And the spam filter is getting clogged up.
According to all reports, academic publishing is drowning in what has come to be called AI slop: low-quality, mass-produced content generated with AI. By 2025, an estimated 57 percent of all published academic articles exhibited detectable evidence of large language model (LLM) influence, a massive jump from 12 percent in 2023. What’s more, over 12 percent of Nature Communications reviews were found to be completely AI-generated in 2025.
Given that for every 40 papers showing statistical evidence of AI usage in 2025, only one formally disclosed it, these figures likely grossly underestimate LLM involvement today. Of course, evidence of AI use doesn’t necessarily mean a paper is slop. But some estimates suggest that hundreds of thousands of fake, often AI-generated, publications are now produced annually.
“It’s not just the [scientific] record under threat, it’s science as a whole,” warns Amanda Sulicz, IEEE’s research-integrity manager.…
The artificial intelligence developer has submitted regulatory paperwork targeting a Nasdaq debut that could value the company at up to $2 trillion next month. Detailed disclosures inside the prospectus reveal robust margins alongside rapid operational scaling, though current revenue remains heavily tethered to a narrow base of heavy technical users. A successful public offering would test public market appetite for capital-intensive frontier labs pursuing long-term transformative capabilities.
The enterprise automation startup has tripled its market worth in rapid succession thanks to strong backing from prominent venture capital investors. Its core platform deploys artificial intelligence agents directly into corporate environments to automate routine IT workflows like credential and permission administration. This massive influx of capital reflects surging institutional confidence in specialized enterprise automation software.
The e-commerce giant is broadening the utility of Alexa for Shopping to handle routine list curation and sensitive post-purchase account tasks. United States customers can now rely on text or voice interactions to manage everyday planning without manually navigating through separate account menus. This push equips the May-launched tool to resolve common customer service friction points directly from the main interface.