OpenAI considers letting users create AI-generated porn

AI Deepfake Porn Scandal Rocks Sydney High School as Police Launch Investigation

But I was curious how security-conscious Meta’s new AI product lineup was, so I decided to see how far I could go. In establishing the commonwealth offence of sharing these images punishable by six years’ imprisonment, the government is adding a companion aggravated offence covering anyone who was also responsible for creating them. Joe Kukura is an SFist staff asst. editor / reporter who has been published in almost every San Francisco publication, including Hoodline, SF Weekly, Thrillist, and Broke Ass Stuart. Maybe you heard about the single most hilarious scandal that happened this weekend.

Flagging the new legislation on Saturday, Dreyfus said the government would not tolerate “this sort of insidious criminal behaviour”. The former Dorian space is coming alive again, and after a significant remodel, it reopened over the weekend as Morella, and Italian restaurant with Argentinian influences. The movie version of Wicked is being released in two parts, with the second part slated for release at Thanksgiving 2025. We’ll see if Mattel’s AI design tool only has a brain by the time that next batch of toys is coming out.

Secret Service Agents Seeking Student Over Trump Video Blocked From School

Online abuse disproportionately targets marginalized groups, but it also often affects women in public spaces, like journalists or politicians. “One of my big concerns is that we’re just going to enter this post-truth world where nothing you see online is trustworthy, because everything can be generated in a very, very realistic yet fake way,” Chowdhury tells me. Taylor says that in the aftermath of that deepfake, she often feels out of control of her own life. She repeatedly beeps her car to make sure that it’s locked, and is often terrified the coffee pot is still on and her house is going to catch on fire. It was the first few months of Covid, and Taylor, who lives with anxiety and obsessive compulsive disorder, says she’d already been facing a general sense of worry and paranoia, but this exacerbated everything.

But the company is particular about how sexually explicit material is described. Last month, the city of San Francisco filed a lawsuit against 18 illegal deepfake websites and apps that offered to undress or “nudify” women and girls. Collectively, the lawsuit said, the sites have been visited over 200 million times in the first six months of 2024.

US Air Force Employee ‘Secretly Took Photos of Kids to Make AI Child Porn Images’

NEW YORK – Ten major tech companies announced Tuesday that they would work together to prevent artificial intelligence from creating and spreading materials depicting child sexual abuse. There are concerns that AI is trained on such materials found online to generate masses of inappropriate images. If the bill passes the House and Senate, it would become the first federal law to protect victims of deepfakes. The recent creation and dissemination of Taylor Swift pornographic deepfakes highlighted the issue of consent and AI-generated sexually explicit content. Studies indicate that approximately 98 per cent of deepfake videos found online are of a pornographic nature, with women being the predominant targets in almost all cases. Meanwhile, the creation and distribution of this material has evolved with new developments in technology.

The Disrupt Explicit Forged Images and Non-Consensual Edits Act of 2024, also known as the DEFIANCE Act, would allow people to sue anyone who creates, shares or receives such imagery. The group collected data in part by developing a custom search engine to find members of the 118th Congress by first and last name, abbreviations or nicknames on 11 well-known deepfake sites. SC stands alone in not criminalizing ‘revenge porn.’ Some legislators hope to change that. She added that she was hopeful that, given the Labour Party’s stated commitment to tackle violence against women, her legislation would be supported in the Commons.

Regulations haven’t caught up to this new kind of sexual abuse

Lawmakers took notice, with members of Congress decrying how deepfakes can inflict “irrevocable emotional, financial, and reputational harm,” disproportionately impacting women. Then, last month, bipartisan legislation dubbed the DEFIANCE Act was introduced to allow victims legal recourse to sue creators and disseminators of non-consensual digital forgeries. Laura Farris, the UK’s minister for victims and safeguarding, called the creation of sexual deepfakes “despicable” and completely “irresponsible”.

Four years later, she recalls that she hadn’t even known what a deepfake was at the time. She opened her boyfriend’s computer and went to the Pornhub link her classmate messaged her. When the website loaded, she saw her face staring back at her in a sexually explicit video she’d never made. A recent piece by Tatum Hunter in the Washington Post examines the future of artificial intelligence in pornography. The industry has always been an early adopter of technology, from the VHS tape to streaming, so it’s no surprise that adult entertainment entrepreneurs are working hard to harness the power of generative AI to make porn. The technology has the potential to disrupt the sprawling and complicated adult entertainment industry, but to what degree?

IWF’s discovery on the dark web centers on a manual that encourages the use of “nudifying” tools to create deepfake child porn and use it against their victims for extortion. This crackdown also follows a disquieting trend of people being victimized by AI-generated pornographic imagery shared widely online without their consent. In January, doctored nudes of Taylor Swift were viewed tens of millions of times across social platforms before being removed, sparking outrage. We’ve seen plenty of examples of people bypassing security filters and limitations placed on generative AI services, including the recent adversarial attacks.

AI Flub May Be Behind Mattel’s ‘Wicked’ Dolls Promoting Porn Website on Packaging – SFist

AI Flub May Be Behind Mattel’s ‘Wicked’ Dolls Promoting Porn Website on Packaging.

Posted: Mon, 11 Nov 2024 08:00:00 GMT [source]

Clare McGlynn, a law professor at Durham University and an expert in pornography regulation, said she questioned any tech company pledge to produce adult content responsibly. Microsoft introduced new protections for its Microsoft Designer product, which uses OpenAI technology, in the wake of the Swift furore this year after a report that it was being used to create unauthorised deepfakes of celebrities. In May, a U.S. man was charged by the FBI for allegedly producing 13,000 sexually explicit and abusive AI images of children on the popular Stable Diffusion model. 42-year-old Steven Anderegg was arrested for creating thousands of “hyper-realistic images of nude and semi-clothed prepubescent children” using generative AI.

“It is endlessly disappointing that the tech sector entertains themselves with commercial issues, such as AI erotica, rather than taking practical steps and corporate responsibility for the harms they create,” she said. Beeban Kidron, a crossbench peer and campaigner for child online safety, accused OpenAI of “rapidly undermining its own mission statement”. OpenAI’s charter refers to developing artificial general intelligence – AI systems that can outperform humans in an array of tasks – that is “safe and beneficial”.

• Collectively, the lawsuit said, the sites have been visited over 200 million times in the first six months of 2024.
• After being contacted by Forbes, TikTok removed the ads for violating its policies.
• Outlawing the sharing of non-consensual deepfake pornographic material was among the commitments arising from a national cabinet meeting on 1 May, at which first ministers pledged themselves to the goal of ending violence against women within a generation.
• The mass production of AI porn has significant ethical and social implications.
• The act focuses on sexually explicit deep fake material, revenge porn and enforcement procedures when it comes to social media platforms removing any content posted.

These fake firms aimed to deceive cybersecurity professionals into working for the criminal organization under the guise of performing penetration testing, instead using them to develop malware and conduct network intrusions. In the early 2000s, cybercriminals began using adult websites to distribute Trojan horses and spyware disguised as video players or codecs. These programs, like the ILOVEYOU virus, recorded keystrokes and changed browser settings without the user’s knowledge. Maddocks has studied how women who speak out in public are more likely to experience digital sexual violence. So surely this Bill is very sensible legislation as part of that mission,” she said. Other legal initiatives to tackle this issue include the Deepfakes Accountability Act, Singapore’s anti-deepfakes legislation, and the EU AI Act.

OpenAI’s “universal policies” require users of its products to “comply with applicable laws” including on exploitation or harm of children, although it does not refer directly to pornographic content. AI models can be used to create this material because they’ve seen examples before. Researchers at Stanford discovered last December that one of the most significant data sets used to train image-generation models included hundreds of pieces of CSAM.

A study by Home Security Heroes found that 94% of individuals featured in deepfake porn are from the entertainment sector, emphasizing the gendered nature of this exploitation. Both bills are in response to developing AI technologies that can use existing photos, videos and recordings to create new depictions — including creating pornography or other obscene materials using someone’s photos or videos posted online. In 2023, the CyberTipline, run by the National Center for Missing and Exploited Children, received 4,700 reports of child sexual abuse material involving generative AI.

OpenAI Is ‘Exploring’ How to Responsibly Generate AI Porn

AI porn typically involves deep learning algorithms and neural networks, particularly Generative Adversarial Networks (GANs). These networks are trained on vast datasets of real pornographic material, learning to generate realistic images and videos. Some AI tools can even create hyper-realistic representations of real people, often without their consent.

• Studies indicate that approximately 98 per cent of deepfake videos found online are of a pornographic nature, with women being the predominant targets in almost all cases.
• More complex text-to-video generators already exist, however, the anticipated release of OpenAI’s model Sora suggests significant progress in text-to-video generation, namely in its high level of realism, complex scene creation and unmatched video length.
• (TikTok’s ad library doesn’t include ads shown to its U.S.-based users) Most of these ads depict celebrities like Scarlett Johansson, Emma Watson and Gal Gadot kissing one another.
• RUMMAN CHOWDHURY IS no stranger to the horrors of online harassment; she was once the head of ethical AI at X, back when it was called Twitter and before Elon Musk decimated her department.
• Beyond images and videos, various sites also allow users to engage with a sex chatbot for conversation.
• Many of the most popular downloadable open-source AI image generators, including the popular Stable Diffusion version 1.5 model, were trained using this data.

When asked to do so, ChatGPT will respond that the request violates its terms of service. As a result, Jang points out that the web is rife with complaints from OpenAI users that the company is censoring them. Dave Willner is a non-resident fellow in the program on the governance of emerging technologies at the Stanford Cyber Policy Center, as well as safety advisor and consultant for tech companies, including OpenAI and Character.AI. He previously worked as head of trust & safety at OpenAI, as head of community policy at Airbnb, and as head of content policy at Meta (formerly Facebook). OpenAI’s models have been trained on vast amounts of public web content, some undoubtedly pornographic in nature.

AI-generated videos of people kissing and hugging are already circulating across social media. A video depicting Taylor Swift hugging Kim Jong Un has about 30 million views on Instagram. In late December, a deepfake video of Elon Musk and Italian Prime Minister Giorgia Meloni kissing went viral on X. “The school has been made aware that a year 12 male student has allegedly used artificial intelligence to create a profile that resembles your daughters and others,” read the school’s email to affected parents, according to local media. “Unfortunately, innocent photos from social media and school events have been used.” Despite slow movement toward fully effective policies, De Mooy said, the more states that introduce and pass laws against explicit deepfakes puts pressure on the federal government to address the issue as well.

Of particular concern is the overwhelming majority of deepfake videos being explicitly pornographic, highlighting the distressing misuse of this technology for generating illicit content. In January alone, MrDeepFakes, a prominent website for deepfake porn, received 88.4 million visits, as reported by Semrush, a US-based online traffic analytics service. The number of synthetic adult videos has seen a dramatic 24-fold increase from 2019 to 2023. Last year, 143,868 new deepfake videos were uploaded online, highlighting the explosive growth of this disturbing trend.

This website is using a security service to protect itself from online attacks. There are several actions that could trigger this block including submitting a certain word or phrase, a SQL command or malformed data. Regardless, maybe what bit Mattel in the ass here is that they thought they were defying gravity by using AI tools to design their toys’ packaging.

In addition, the images are often impossible to completely erase from the internet, ensuring continued trauma. And the market is truly transnational; a 2022 sting by New Zealand found a network of child-pornography sharers across 12 countries. A U.S. Air Force employee was arrested for secretly taking photos of children in order to create AI child abuse images. Anderegg also allegedly kept in contact with a 15-year-old boy and told him how he used Stable Diffusion to convert text prompts into child sex abuse images, according to the Justice Department.