As the fight against deepfakes<\/a> heats up, one company is helping us fight back. Hugging Face<\/a>, a company that hosts AI projects and machine learning<\/a> tools has developed a range of\u00a0 \u201cstate of the art technology\u201d to combat \u201cthe rise of AI-generated \u2018fake\u2019 human content\u201d like deepfakes and voice scams.<\/p>\n This range of technology includes a collection of tools labeled \u2018Provenance, Watermarking and Deepfake Detection.\u2019 There are tools that not only detect deepfakes<\/a> but also help by embedding watermarks in audio files, LLMs, and images.<\/p>\n Margaret Mitchell, researcher and chief ethics scientist<\/a> at Hugging Face, announced the tools in a lengthy Twitter thread, where she broke down how each of these different tools works. The audio watermarking tool, for instance, works by embedding an\u00a0 \u201cimperceptible signal that can be used to identify synthetic voices as fake,\u201d while the image \u201cpoisoning\u201d tool works by \u201cdisrupt[ing] the ability to create facial recognition models.\u201d<\/p>\n Furthermore, the image \u201cguarding\u201d tool, Photoguard, works by making an image \u201cimmune\u201d to direct editing by generative models. There are also tools like Fawkes, which work by limiting the use of facial recognition software on pictures that are accessible publicly, and numerous embedding tools that work by embedding watermarks that can be detected by specific software. Such embedding tools include Imatag, WaveMark, and Truepic.<\/p>\n With the rise of AI-generated “fake” human content\u2013”deepfake” imagery, voice cloning scams & chatbot babble plagiarism\u2013those of us working on social impact @huggingface<\/a> put together a collection of some of the state-of-the-art technology that can help:https:\/\/t.co\/nFS7GW8dtk<\/a><\/p>\n \u2014 MMitchell (@mmitchell_ai) February 12, 2024<\/a><\/p>\n<\/blockquote>\n While these tools are certainly a good start, Mashable <\/a>tech reporter Cecily Mauran warned there might be some limitations. \u201cAdding watermarks<\/a> to media created by generative AI is becoming critical for the protection of creative works and the identification of misleading information, but it\u2019s not foolproof,\u201d she explains in an article for the outlet. \u201cWatermarks embedded within metadata are often automatically removed when uploaded to third-party sites like social media, and nefarious users can find workarounds by taking a screenshot of a watermarked image.\u201d<\/p>\n \u201cNonetheless,\u201d she adds, \u201cfree and available tools like the ones Hugging Face shared are way better than nothing.\u201d<\/p>\n Featured Image: Photo by Vishnu Mohanan<\/a> on Unsplash<\/a><\/p>\nIntroducing Hugging Face<\/h2>\n
\n
![\"\"](\"https:\/\/secure.gravatar.com\/avatar\/afeaa6bd1b8144e682f922a0727700e3?s=125&d=mm&r=g\")