Deepfakes
Ongoing coverage on Deepfakes.
Latest coverage
That Heartwarming Animal Clip in Your Feed? Assume It’s Fake [2026]
The proliferation of AI-generated images and deepfakes is making it increasingly difficult to discern real animal footage from fabricated content, leading to scams and eroding trust in legitimate animal welfare organizations. While new laws and AI features aim to identify machine-made content, current solutions are partial, leaving individuals like Mibbby Butler vulnerable to scams involving lost pets.
They Built Careers on Trust in the Classroom. Then Students Started Making Deepfakes of Them
Educators are facing a disturbing trend of students creating and distributing AI-generated deepfakes, often sexualized, of their teachers. This misuse of artificial intelligence is causing significant distress and impacting the trust within classrooms. Schools are struggling to address these incidents, highlighting a lack of clear protocols for handling cyberbullying and the misuse of AI technology in educational settings.
Interpol Flags AI as the Engine Behind Africa’s Doubling Cybercrime Losses
Interpol's latest report indicates that cybercrime losses in Africa have more than doubled, reaching $484 million, with Artificial Intelligence identified as a key driver in 55 percent of reported cases. AI is now considered the core operational driver of cybercrime on the continent, automating all stages of attacks from reconnaissance to evasion, despite enforcement efforts leading to over 1,500 arrests and significant seizures.
Judge Lets Minnesota’s Nudification Law Stand as xAI’s Last-Minute Block Attempt Fails
A federal judge has allowed Minnesota's law against nonconsensual intimate imagery to stand, rejecting a last-minute attempt by xAI to block it. The judge cited xAI's significant delay in challenging the law, which imposes a $500,000 penalty on developers for each instance of misuse. While the company argues the law infringes on free speech, litigation is ongoing, with a hearing scheduled for August 19 to decide on a preliminary injunction.
Hugging Face reportedly plagued with AI models generating adult deepfakes
Researchers have found that a significant number of AI models hosted on Hugging Face are capable of generating non-consensual deepfakes, including adult content. This discovery raises concerns about the misuse of AI technology and content moderation on the platform.
Why new EU deepfake legislation means games need to be careful when featuring real-world people and locations
New EU AI Act legislation, effective August 2, 2026, introduces transparency obligations for AI-generated and manipulated content, including deepfakes. While common game development uses of AI like fictional character voices or environments are unlikely to be classified as deepfakes, the act raises concerns for content featuring realistic depictions of real people, places, or events, and its promotional materials. Game studios must develop tailored AI content policies to ensure compliance with labeling requirements, especially for content distributed within the EU.
xAI sues Grok user for generating nonconsensual sexualized deepfakes
xAI has initiated legal action against an individual accused of using its AI tool, Grok, to create nonconsensual sexualized deepfakes, including images of minors. The lawsuit highlights concerns surrounding the misuse of AI for generating harmful content.
Meta deactivates feature that let you generate AI images of any public Instagram account
Meta has deactivated its Muse Image feature, which allowed users to generate AI images of public Instagram accounts. The capability has been removed following its misuse.
The Oversight Board says Meta needs to do more to protect regular people from sexualized deepfakes
Meta's Oversight Board has recommended improvements to the reporting process for sexualized AI deepfakes, aiming to make it easier and more effective for individuals to flag such content.
Ubisoft closing studios in Winnipeg and Belgrade and cutting publishing roles worldwide, 380 jobs at risk
Ubisoft is closing its studios in Winnipeg and Belgrade and making further cuts to its global publishing team, affecting approximately 380 roles. The Barcelona studio will refocus exclusively on the Rainbow Six franchise. Separately, the Entertainment Software Association is warning that a proposed 'No Fakes Act' could harm the video game industry by failing to distinguish between harmful deepfakes and legitimate digital replicas used in game development.
ESA says anti-deepfake act could 'devastate' video game industry
The Entertainment Software Association (ESA) has expressed strong opposition to SAG-AFTRA's proposed "No Fakes" act, arguing that its current wording creates significant legal uncertainty and could devastate the video game industry. ESA president Stanley Pierre-Louis stated that the act's broad definition of "digital replica" could lead to frivolous lawsuits and that the cost of litigation would be economically devastating, despite the industry likely prevailing in court.
ESA warns 'No Fakes Act' fails to distinguish between deepfakes and digital replicas used in video games
The Entertainment Software Association (ESA) has warned that the proposed 'No Fakes Act' could negatively impact the video game industry by failing to distinguish between harmful deepfakes and legitimate digital replicas used in game development. ESA president Stanley Pierre-Louis urged the US Senate Judiciary Committee to amend the bill, expressing concerns that its current definition could lead to frivolous lawsuits and economically devastating litigation for developers.
Evening Legal Briefs
xAI has requested a court to reveal the identities of individuals who allegedly created and distributed deepfake nude images of victims, which were reportedly used in connection with its Grok AI. Separately, a Grand Theft Auto 6 YouTuber was detained after attempting to trespass onto Rockstar Games property.
Meta's Oversight Board will examine deepfake video of a UK politician Facebook left online
Meta's Oversight Board will review a deepfake video of a UK politician that was left online. The video, which appears to be AI-generated, shows the politician making controversial remarks about refugees.
YouTube to expand likeness detection to all creators 18 & over
YouTube is expanding its likeness detection feature to all creators aged 18 and over. This move by Google aims to combat misinformation and protect users in response to the rise of AI and deepfakes.
Morning Safety Dance
The article discusses the growing problem of deepfakes and their impact on eroding trust. It emphasizes the critical need for verification tools to combat misinformation and restore confidence in digital content.
Evening Legal Briefs
Pop star Taylor Swift is seeking to trademark her likeness, a move highlighted by the emergence of TikTok advertisements using deepfake technology. This situation raises questions about intellectual property protection in the digital age and the potential misuse of celebrity images.
Apple and Google are reportedly pointing users to nudify apps
Apple and Google are reportedly continuing to promote "nudify" apps on their respective app stores, despite policies against such content. These apps, which can create explicit images of real people, have generated significant revenue and downloads, with some even rated for children. Both companies claim to be taking action, but a report suggests they are actively directing users to these problematic applications.
Evening Legal Briefs
Meta has been ordered by a jury to pay $375 million for violating New Mexico law in a child exploitation case. Separately, Baltimore is suing xAI concerning alleged deepfakes generated by its Grok AI.
Baltimore sues xAI over Grok deepfakes
The city of Baltimore has filed a lawsuit against xAI, alleging that its AI chatbot Grok violated the city's Consumer Protection Ordinance. This action follows widespread criticism of Grok's image generation tool for creating millions of sexualized images, including those of minors. The lawsuit claims xAI failed to disclose the risks associated with using Grok and the X social network.