Find similar content

#deepfakes#book 📚 Exploring Deepfakes (2023) 🔗Link ----- Main channel:@repo_science Coupons: @freecoupons_reposcience -----

🇬🇧Ofcom to Platforms: Deepfake Detection Is Not the User’s Job The U.K. Office of Communications (Ofcom) has released a detailed paper outlining how individuals might identify AI-generated deepfakesь but with a clear message: this responsibility should not fall solely on users. The report, grounded in expert interviews, user research, literature review, and technical testing of open-source watermarking tools, outlines eight key findings. Ofcom emphasizes that platforms must take proactive responsibility. Relying on individuals to detect synthetic media risks normalizing misinformation and erodes public trust. Technical safeguards like robust watermarking and platform-level interventions are not optional; they’re governance essentials in the age of generative AI. #AI#Deepfakes

US: Gym Teacher Allegedly Used AI Voice Clone to Frame Principal in Racist Rant Dazhon Darien, a former athletic director at Pikesville High School, is accused of creating a fake audio recording featuring the voice of Principal Eric Eiswert spewing racist and antisemitic remarks. The recording circulated on social media, leading to Eiswert's temporary suspension. Investigators believe Darien employed generative AI, a technology capable of replicating someone's voice with remarkable accuracy. Experts identified inconsistencies in the recording, such as a flat tone and lack of natural background noise, suggesting manipulation. Authorities linked the recording back to Darien by tracing the access and use of school computers to access AI tools like OpenAI and Microsoft Bing Chat. Additional evidence connected Darien to the audio's release through an email address and linked phone number. Police believe Darien fabricated the recording in retaliation for an investigation into potential mishandling of school funds, in which he was allegedly implicated. Darien was arrested at the airport and charged with theft, disrupting school operations, retaliation, and stalking. He has been released on bail. This incident highlights the growing capabilities and potential dangers of AI voice cloning technology. #AI#Deepfakes

Florida teens accused of creating deepfake nudes of their classmates Hello everyone! Two Miami teenagers have been arrested for creating and distributing AI-generated nude photos of their classmates without their consent. The boys, aged 13 and 14, are facing criminal charges and the parents of the victims are demanding that the offenders be expelled from school. With the advancement of AI image creation technology, it is becoming increasingly difficult to combat its abuse. It is a serious breach of personal privacy and can have psychological consequences for victims. #AI#Deepfakes

Stanford Professor Accused of Using Fake AI Citations in Deepfake Bill Support Stanford Professor Jeff Hancock, a prominent figure in misinformation research, is under scrutiny for allegedly including AI-generated fake citations in a legal argument supporting Minnesota’s proposed deepfake legislation. The law seeks to regulate deepfakes during elections, but questions about the credibility of Hancock’s filing have sparked controversy. Key citations, such as the study “Deepfakes and the Illusion of Authenticity,” appear unverifiable, leading critics to suspect AI involvement. Opponents argue that these “AI hallucinations” undermine the reliability of the entire argument. Representative Mary Franson stated the filing’s credibility is now in question. This incident highlights a critical challenge in integrating AI into academic and legal work: ensuring transparency and accountability in generated content. As Hancock’s silence continues, the debate over the ethical use of AI in shaping public policy intensifies. #AI#Deepfakes#AIEthics

🇺🇸Judge Rejects AI-Generated Citations in Minnesota Deepfake Case A federal judge in Minnesota excluded expert testimony from misinformation specialist Jeff Hancock after discovering that his court filing contained AI-generated fake citations. Hancock, a Stanford professor, admitted to using ChatGPT-4o, which likely "hallucinated" two citations. The case involves a Minnesota law banning AI-generated deepfakes in elections, defended by Attorney General Keith Ellison. Judge Laura Provinzino stated that while Hancock did not intentionally use fake sources, this mistake "shatters his credibility with this court." She prohibited Ellison from submitting revised testimony from Hancock but declined to block the deepfake law. The case highlights growing concerns over AI-generated misinformation in legal proceedings. #AIRegulation#Deepfakes#LegalTech

French prosecutors summon Elon Musk over allegations of child abuse images and deepfakes on X [Read FullArticle] @WorldNews#ElonMusk#Deepfakes#WorldNews

UK Cracks Down on Creators of AI Sex Deepfakes The UK government is taking a strong stance against the creation of AI-generated sexually explicit deepfakes. A new law, currently in its proposal stage, aims to criminalize the creation of these non-consensual images, even if they are not widely distributed. Deepfake technology allows for the creation of realistic pornography by superimposing a person's face onto another body. This practice has become increasingly accessible due to advancements in AI image synthesis tools. The UK government sees this as a form of harassment and a violation of privacy, particularly for women. Under the new legislation, creating a deepfake sexual image without consent would be considered a criminal offense, regardless of the intent to share it. Offenders could face: ✅Unlimited fines: Highlighting the severity of the offense. ✅Criminal record: Potentially impacting future employment and opportunities. Additionally, sharing the created deepfake would potentially result in jail time. This law builds upon the Online Safety Act passed last year, which criminalized sharing non-consensual deepfakes. This new proposal tackles the root of the problem, targeting the creation of the content itself. The UK isn't alone in addressing this issue. Recent cases like the creation of deepfake nudes of middle school girls in Florida highlight the real-world impact of this technology. The proposed law is not yet enacted and needs parliamentary approval. The government is also looking to strengthen existing laws to allow harsher penalties for both creating and distributing deepfakes. #AI#Deepfakes#Cybersecurity

🇺🇸AI-generated Evidence is Showing up in Court A California court has confronted one of the clearest examples of AI-generated evidence being submitted as if it were real. In &Mendones v. Cushman & Wakefield, Inc. Judge Victoria Kolakowski identified a witness video (Exhibit 6C) with distorted voice patterns, repetitive facial movements, and blurred expressions. These anomalies led her to conclude that the submission was an AI-generated deepfake presented as authentic testimony. On September 9, 2025, the Alameda County Superior Court issued a terminating sanction, dismissing the case with prejudice due to the falsified evidence, and later denied the plaintiffs’ request for reconsideration. #AI#Deepfakes#AIGovernance#LegalTech

Tackling AI-Generated Deepfakes: Insights from NIST The U.S. National Institute of Standards and Technology (NIST) has released a pivotal report, "Reducing Risks Posed by Synthetic Content" offering actionable strategies against the misuse of AI-powered deepfakes. The report centers on three critical measures: tracking the provenance of content to reveal its origins and changes, labeling AI-generated content, and combatting the production and spread of AI-generated CSAM and non-consensual intimate images (NCII). NIST highlights the multifaceted risks posed by synthetic content, including disinformation affecting society at large and cybersecurity threats like voice cloning for biometric fraud. It emphasizes the importance of provenance tracking to ensure transparency and tools for analysts to detect AI-generated content. However, the report also cautions that even authentic content can be harmful, reminding stakeholders to approach synthetic content risks with nuance and precision. #Deepfakes#AIEthics#CyberSecurity

🇮🇹Italian DPA Bans Deep Nude App Over Human Dignity and Privacy Risks Garante per la protezione dei dati personali (Italian DPA) has officially banned the app Clothoff, which enables users to generate “deep nude” images of real people, including minors, without consent or verification. The authority highlighted that the app allows both free and paid creation of fake explicit content without any disclosure that the images are AI-generated. The DPA emphasized that the ban was necessary due to “high risks to fundamental rights and freedoms,” particularly concerning human dignity, privacy, and personal data protection. An investigation has been launched to address the broader ecosystem of nudity apps. Recent Italian news reports show how the misuse of such generative tools has already caused significant social alarm. #AIRegulation#Ethics#Deepfakes

🇺🇸US Senate Passes AI Deepfake Bill The US Senate has approved the "Take It Down Act", a bipartisan bill co-sponsored by Sens. Ted Cruz and Amy Klobuchar. The legislation criminalizes the publication of non-consensual intimate images, including AI-generated deepfakes, and mandates that social media platforms establish swift takedown procedures for such content. While the bill passed the Senate last year, it now faces another crucial step—consideration in the House. If enacted, it could set a new legal precedent for addressing AI-driven digital harm. #AIRegulation#Deepfakes#AIethics#TechPolicy