Post
233
Are you sure the open-source LLM model you just downloaded is safe?
A recent paper on "Privacy Backdoors" reports a new vulnerability where pre-trained models can be poisoned before fine-tuning them. This is a serious challenge for everyone building on open-source AI.
Instead of just pointing out problems, we believe in finding better solutions. To understand this threat, the researchers needed to test their attack on realistic data structures. They needed a dataset that could effectively simulate a high-stakes privacy attack, and we're proud that our Ai4Privacy dataset was used to provide this crucial benchmark. The paper reports that for our complex dataset, the privacy leakage on a non-poisoned model was almost zero. After the backdoor attack, that number reportedly jumped to 87%.
Ai4Privacy dataset provided a realistic benchmark for their research. Our dataset, composed of synthetic identities, helped them demonstrate how a poisoned model could dramatically amplify privacy leakage.
This is why we champion open source: it enables the community to identify these issues and develop better, safer solutions together.
Kudos to the research team behind this study: Yuxin Wen, Leo Marchyok, Sanghyun Hong, Jonas Geiping, Tom Goldstein, and Nicholas Carlini, Oregon State University, University of Maryland, Google DeepMind, and ELLIS Institute Tubingen & MPI Intelligent Systems.
🔗 Read the research to understand this new challenge: https://arxiv.org/pdf/2404.01231
#DataPrivacy #AI #OpenSource #Anonymization #MachineLearning #Ai4Privacy #Worldslargestopensourceprivacydataset
A recent paper on "Privacy Backdoors" reports a new vulnerability where pre-trained models can be poisoned before fine-tuning them. This is a serious challenge for everyone building on open-source AI.
Instead of just pointing out problems, we believe in finding better solutions. To understand this threat, the researchers needed to test their attack on realistic data structures. They needed a dataset that could effectively simulate a high-stakes privacy attack, and we're proud that our Ai4Privacy dataset was used to provide this crucial benchmark. The paper reports that for our complex dataset, the privacy leakage on a non-poisoned model was almost zero. After the backdoor attack, that number reportedly jumped to 87%.
Ai4Privacy dataset provided a realistic benchmark for their research. Our dataset, composed of synthetic identities, helped them demonstrate how a poisoned model could dramatically amplify privacy leakage.
This is why we champion open source: it enables the community to identify these issues and develop better, safer solutions together.
Kudos to the research team behind this study: Yuxin Wen, Leo Marchyok, Sanghyun Hong, Jonas Geiping, Tom Goldstein, and Nicholas Carlini, Oregon State University, University of Maryland, Google DeepMind, and ELLIS Institute Tubingen & MPI Intelligent Systems.
🔗 Read the research to understand this new challenge: https://arxiv.org/pdf/2404.01231
#DataPrivacy #AI #OpenSource #Anonymization #MachineLearning #Ai4Privacy #Worldslargestopensourceprivacydataset