{"slug": "mit-s-new-method-flags-ai-models-trained-on-casm-without-generating-it", "title": "MIT's New Method Flags AI Models Trained on CASM Without Generating It", "summary": "MIT researchers developed a new auditing method called Gaussian probing that can detect AI models fine-tuned to generate child sexual abuse material (CSAM) without producing any images, achieving 100% accuracy. The technique, created in collaboration with child safety nonprofit Thorn, analyzes internal model adaptations rather than outputs, bypassing legal and ethical barriers. This breakthrough addresses a critical blind spot as AI-generated CSAM reports surged from 67,000 in 2024 to over 1.5 million in 2025.", "body_md": "**July 13, 2026**, (Inside AI) — A new auditing method developed by MIT researchers can determine whether an AI model has been fine-tuned to generate child sexual abuse material (CSAM) without ever producing an image, sidestepping the legal and ethical barriers that have stymied safety checks.\n\nLed by graduate student **Vinith Suriyakumar** and associate professors **Ashia Wilson** and **Marzyeh Ghassemi**, the team collaborated with child safety nonprofit **Thorn** to create a technique that inspects a model’s internal adaptations rather than its outputs. The approach, detailed in a paper presented at the **International Conference on Machine Learning**, achieved **100% accuracy** in identifying models specialized for CSAM generation.\n\nThe breakthrough comes as reports of AI-generated CSAM skyrocket. The **National Center for Missing and Exploited Children** fielded over **1.5 million** such reports in **2025**, up from just **67,000** in **2024**. Open-source generative AI models, easily fine-tuned using methods like **low-rank adaptation (LoRA)**, have enabled bad actors to create hyper-realistic abusive imagery at scale.\n\n“This unlocks a new avenue for platforms that host open-source models and for law enforcement to actually test whether a model is capable of generating CSAM. Before, we had no way of measuring this. It was a huge blind spot that some people were taking advantage of. Now, we can address an AI safety problem that is having severe negative impacts,” Suriyakumar said.\n\n## Why Output-Based Audits Failed\n\nTraditional safety testing involves prompting a model and examining its responses. But generating CSAM, even for testing, is illegal in the U.S. and many other jurisdictions. This created a paradox: auditors couldn’t verify if a model was dangerous without committing a crime. Manual checks also don’t scale to the thousands of model variants uploaded monthly, and exposing human reviewers to such content carries psychological risks.\n\nThe MIT-Thorn team bypassed generation entirely. Their method, called **Gaussian probing**, feeds random data points into a model and analyzes how its internal representations shift due to LoRA adaptors—the lightweight add-ons that specialize a base model. By capturing these shifts at multiple layers and averaging them, the probe creates a fingerprint of the adaptation’s purpose.\n\n“We never run the model all the way to the end or prompt the model, so we never generate images,” Suriyakumar explained. The technique proved robust: when tested on variations of three model types, it correctly flagged CSAM-tuned versions every time, even distinguishing them from models fine-tuned for other harmful but non-CSAM content.\n\n## A Scalable Shield for Hosting Platforms\n\nBecause Gaussian probing requires no image generation and minimal computation, it can be integrated into model-hosting platforms like **Hugging Face** or **Civitai** to automatically screen uploads. This could stop dangerous models before they spread, addressing a gap that has allowed illicit LoRA adaptors to proliferate on public repositories.\n\nThe approach also resists evasion better than output filters. A malicious actor would need to fundamentally alter the base model’s architecture to hide the telltale adaptations, a far higher bar than simply tweaking prompts. “There is a huge bucket of child safety concerns with AI, and these are real concerns that need to be addressed. A lot of children are being harmed by AI deepfakes. We’ve shown that Gaussian probing can be a very useful tool, and we hope the research community really pours more attention into this problem,” Wilson said.\n\nThe research, supported in part by the **Bridgewater AIA Labs Research Fellowship**, also involved **Lena Stempfle**, an MIT postdoc, and collaborators from **Boston University** and Thorn. The team plans to test the method on a broader range of models and explore whether it can detect harmful capabilities in base models before any fine-tuning occurs—a preemptive strike against misuse.\n\nWhile the technique marks a significant step, experts caution that it addresses only one vector of AI-generated CSAM. Models trained from scratch on abusive datasets or those using other adaptation methods may still evade detection. Nonetheless, for the open-source ecosystem where LoRA has become the de facto customization tool, Gaussian probing offers a pragmatic, legally sound solution.\n\n“Now we have a technological approach to partially address this concern. So much effort was poured into this collaboration, which enabled us to tackle a really hard problem that is harming so many children, nationally and around the world. Hopefully, we can have a transformative impact in this area,” Ghassemi said.", "url": "https://wpnews.pro/news/mit-s-new-method-flags-ai-models-trained-on-casm-without-generating-it", "canonical_source": "https://insideai.news/news/ai-safety/mits-new-method-flags-ai-models-trained-on-child-abuse-imagery-without-generating-it/3869/", "published_at": "2026-07-13 21:44:18+00:00", "updated_at": "2026-07-13 22:05:26.362282+00:00", "lang": "en", "topics": ["ai-safety", "ai-ethics", "machine-learning"], "entities": ["MIT", "Thorn", "Vinith Suriyakumar", "Ashia Wilson", "Marzyeh Ghassemi", "National Center for Missing and Exploited Children", "Hugging Face", "Civitai"], "alternates": {"html": "https://wpnews.pro/news/mit-s-new-method-flags-ai-models-trained-on-casm-without-generating-it", "markdown": "https://wpnews.pro/news/mit-s-new-method-flags-ai-models-trained-on-casm-without-generating-it.md", "text": "https://wpnews.pro/news/mit-s-new-method-flags-ai-models-trained-on-casm-without-generating-it.txt", "jsonld": "https://wpnews.pro/news/mit-s-new-method-flags-ai-models-trained-on-casm-without-generating-it.jsonld"}}