Архив метки: Stable Diffusion

QuickVid uses AI to generate short-form videos, complete with voiceovers

Generative AI is coming for videos. A new website, QuickVid, combines several generative AI systems into a single tool for automatically creating short-form YouTube, Instagram, TikTok and Snapchat videos.
Given as little as a single word, QuickVid chooses a background video from a library, writes a script and keywords, overlays images generated by DALL-E 2 and adds a synthetic voiceover and background music from YouTube’s royalty-free music library. QuickVid’s creator, Daniel Habib, says that he’s building the service to help creators meet the “ever-growing” demand from their fans.
“By providing creators with tools to quickly and easily produce quality content, QuickVid helps creators increase their content output, reducing the risk of burnout,” Habib told TechCrunch in an email interview. “Our goal is to empower your favorite creator to keep up with the demands of their audience by leveraging advancements in AI.”
But depending on how they’re used, tools like QuickVid threaten to flood already-crowded channels with spammy and duplicative content. They also face potential backlash from creators who opt not to use the tools, whether because of cost ($10 per month) or on principle, yet might have to compete with a raft of new AI-generated videos.
Going after video
QuickVid, which Habib, a self-taught developer who previously worked at Meta on Facebook Live and video infrastructure, built in a matter of weeks, launched on December 27. It’s relatively bare bones at present — Habib says that more personalization options will arrive in January — but QuickVid can cobble together the components that make up a typical informational YouTube Short or TikTok video, including captions and even avatars.
It’s easy to use. First, a user enters a prompt describing the subject matter of the video they want to create. QuickVid uses the prompt to generate a script, leveraging the generative text powers of GPT-3. From keywords either extracted from the script automatically or entered manually, QuickVid selects a background video from the royalty-free stock media library Pexels and generates overlay images using DALL-E 2. It then outputs a voiceover via Google Cloud’s text-to-speech API — Habib says that users will soon be able to clone their voice — before combining all these elements into a video.
Image Credits: QuickVid
See this video made with the prompt “Cats”:

https://techcrunch.com/wp-content/uploads/2022/12/img_5pg7k95x9ig2tofh7mkrr_cfr.mp4
Or this one:
https://techcrunch.com/wp-content/uploads/2022/12/img_61ighv4x55slq9582dbx_cfr.mp4
QuickVid certainly isn’t pushing the boundaries of what’s possible with generative AI. Both Meta and Google have showcased AI systems that can generate completely original clips given a text prompt. But QuickVid amalgamates existing AI to exploit the repetitive, templated format of B-roll-heavy short-form videos, getting around the problem of having to generate the footage itself.
“Successful creators have an extremely high-quality bar and aren’t interested in putting out content that they don’t feel is in their own voice,” Habib said. “This is the use case we’re focused on.”
That supposedly being the case, in terms of quality, QuickVid’s videos are generally a mixed bag. The background videos tend to be a bit random or only tangentially related to the topic, which isn’t surprising given QuickVids being currently limited to the Pexels catalog. The DALL-E 2-generated images, meanwhile, exhibit the limitations of today’s text-to-image tech, like garbled text and off proportions.
In response to my feedback, Habib said that QuickVid is “being tested and tinkered with daily.”
Copyright issues
According to Habib, QuickVid users retain the right to use the content they create commercially and have permission to monetize it on platforms like YouTube. But the copyright status around AI-generated content is … nebulous, at least presently. The U.S. Patent and Trademark Office (USPTO) recently moved to revoke copyright protection for an AI-generated comic, for example, saying copyrightable works require human authorship.
When asked about how the USPTO decision might affect QuickVid, Habib said he believes that it only pertain to the “patentability” of AI-generated products and not the rights of creators to use and monetize their content. Creators, he pointed out, aren’t often submitting patents for videos and usually lean into the creator economy, letting other creators repurpose their clips to increase their own reach.
“Creators care about putting out high-quality content in their voice that will help grow their channel,” Habib said.
Another legal challenge on the horizon might affect QuickVid’s DALL-E 2 integration — and, by extension, the site’s ability to generate image overlays. Microsoft, GitHub and OpenAI are being sued in a class action lawsuit that accuses them of violating copyright law by allowing Copilot, a code-generating system, to regurgitate sections of licensed code without providing credit. (Copilot was co-developed by OpenAI and GitHub, which Microsoft owns.) The case has implications for generative art AI like DALL-E 2, which similarly has been found to copy and paste from the datasets on which they were trained (i.e., images).
Habib isn’t concerned, arguing that the generative AI genie’s out of the bottle. “If another lawsuit showed up and OpenAI disappeared tomorrow, there are several alternatives that could power QuickVid,” he said, referring to the open source DALL-E 2-like system Stable Diffusion. QuickVid is already testing Stable Diffusion for generating avatar pics.
Moderation and spam
Aside from the legal dilemmas, QuickVid might soon have a moderation problem on its hands. While OpenAI has implemented filters and techniques to prevent them, generative AI has well-known toxicity and factual accuracy problems. GPT-3 spouts misinformation, particularly about recent events, which are beyond the boundaries of its knowledge base. And ChatGPT, a fine-tuned offspring of GPT-3, has been shown to use sexist and racist language.
That’s worrisome, particularly for people who’d use QuickVid to create informational videos. In a quick test, I had my partner — who’s far more creative than me, particularly in this area —  enter a few offensive prompts to see what QuickVid would generate. To QuickVid’s credit, obviously problematic prompts like “Jewish new world order” and “9/11 conspiracy theory” didn’t yield toxic scripts. But for “Critical race theory indoctrinating students,” QuickVid generated a video implying that critical race theory could be used to brainwash schoolchildren.
See:

https://techcrunch.com/wp-content/uploads/2022/12/img_e4wba39us0vqtc8051491_cfr.mp4
Habib says that he’s relying on OpenAI’s filters to do most of the moderation work and asserts that it’s incumbent on users to manually review every video created by QuickVid to ensure “everything is within the boundaries of the law.”
“As a general rule, I believe people should be able to express themselves and create whatever content they want,” Habib said.
That apparently includes spammy content. Habib makes the case that the video platforms’ algorithms, not QuickVid, are best positioned to determine the quality of a video, and that people who produce low-quality content “are only damaging their own reputations.” The reputational damage will naturally disincentivize people from creating mass spam campaigns with QuickVid, he says.
“If people don’t want to watch your video, then you won’t receive distribution on platforms like YouTube,” he added. “Producing low-quality content will also make people look at your channel in a negative light.”
But it’s instructive to look at ad agencies like Fractl, which in 2019 used an AI system called Grover to generate an entire site of marketing materials — reputation be damned. In an interview with The Verge, Fractl partner Kristin Tynski said that she foresaw generative AI enabling “a massive tsunami of computer-generated content across every niche imaginable.”
In any case, video-sharing platforms like TikTok and YouTube haven’t had to contend with moderating AI-generated content on a massive scale. Deepfakes — synthetic videos that replace an existing person with someone else’s likeness — began to populate platforms like YouTube several years ago, driven by tools that made deepfaked footage easier to produce. But unlike even the most convincing deepfakes today, the types of videos QuickVid creates aren’t obviously AI-generated in any way.
Google Search’s policy on AI-generated text might be a preview of what’s to come in the video domain. Google doesn’t treat synthetic text differently from human-written text where it concerns search rankings but takes actions on content that’s “intended to manipulate search rankings and not help users.” That includes content stitched together or combined from different web pages that “[doesn’t] add sufficient value” as well as content generated through purely automated processes, both of which might apply to QuickVid.
In other words, AI-generated videos might not be banned from platforms outright should they take off in a major way but rather simply become the cost of doing business. That isn’t likely to allay the fears of experts who believe that platforms like TikTok are becoming a new home for misleading videos, but — as Habib said during the interview — “there is no stopping the generative AI revolution.”
QuickVid uses AI to generate short-form videos, complete with voiceovers by Kyle Wiggers originally published on TechCrunch
QuickVid uses AI to generate short-form videos, complete with voiceovers

OpenAI begins allowing users to edit faces with DALL-E 2

After initially disabling the capability, OpenAI today announced that customers with access to DALL-E 2 can upload people’s faces to edit them using the AI-powered image-generating system. Previously, OpenAI only allowed users to work with and share photorealistic faces and banned the uploading of any photo that might depict a real person, including photos of prominent celebrities and public figures.
OpenAI claims that improvements to its safety system made the face-editing feature possible by “minimizing the potential of harm” from deepfakes as well as attempts to create sexual, political and violent content. In an email to customers, the company wrote:
Many of you have told us that you miss using DALL-E to dream up outfits and hairstyles on yourselves and edit the backgrounds of family photos. A reconstructive surgeon told us that he’d been using DALL-E to help his patients visualize results. And filmmakers have told us that they want to be able to edit images of scenes with people to help speed up their creative processes … [We] built new detection and response techniques to stop misuse.
The change in policy isn’t opening the floodgates necessarily. OpenAI’s terms of service will continue to prohibit uploading pictures of people without their consent or images that users don’t have the rights to — although it’s not clear how consistent the company’s historically been about enforcing those policies.
In any case, it’ll be a true test of OpenAI’s filtering technology, which some customers in the past have complained about being overzealous and somewhat inaccurate. Deepfakes come in many flavors, from fake vacation photos to presidents of war-torn countries. Accounting for every emerging form of abuse will be a never-ending battle, in some cases with very high stakes.
No doubt, OpenAI — which has the backing of Microsoft and notable VC firms including Khosla Ventures — is eager to avoid the controversy associated with Stability AI’s Stable Diffusion, an image-generating system that’s available in an open source format without any restrictions. As TechCrunch recently wrote about, it didn’t take long before Stable Diffusion — which can also edit face images — was being used by some to create pornographic, nonconsensual deepfakes of celebrities like Emma Watson.
So far, OpenAI has positioned itself as a brand-friendly, buttoned-up alternative to the no-holds-barred Stability AI. And with the constraints around the new face editing feature for DALL-E 2, the company is maintaining the status quo.
DALL-E 2 remains in invite-only beta. In late August, OpenAI announced that over a million people are using the service.
OpenAI begins allowing users to edit faces with DALL-E 2 by Kyle Wiggers originally published on TechCrunch
OpenAI begins allowing users to edit faces with DALL-E 2

AI is getting better at generating porn. We might not be prepared for the consequences.

A red-headed woman stands on the moon, her face obscured. Her naked body looks like it belongs on a poster you’d find on a hormonal teenager’s bedroom wall — that is, until you reach her torso, where three arms spit out of her shoulders.
AI-powered systems like Stable Diffusion, which translate text prompts into pictures, have been used by brands and artists to create concept images, award-winning (albeit controversial) prints and full-blown marketing campaigns.
But some users, intent on exploring the systems’ murkier side, have been testing them for a different sort of use case: porn.
AI porn is about as unsettling and imperfect as you’d expect (that red-head on the moon was likely not generated by someone with an extra arm fetish). But as the tech continues to improve, it will evoke challenging questions for AI ethicists and sex workers alike.
Pornography created using the latest image-generating systems first arrived on the scene via the discussion boards 4chan and Reddit earlier this month, after a member of 4chan leaked the open source Stable Diffusion system ahead of its official release. Then, last week, what appears to be one of the first websites dedicated to high-fidelity AI porn generation launched.
Called Porn Pen, the website allows users to customize the appearance of nude AI-generated models — all of which are women — using toggleable tags like “babe,” “lingerie model,” “chubby,” ethnicities (e.g. “Russian” and “Latina”) and backdrops (e.g. “bedroom,” “shower” and wildcards like “moon”). Buttons capture models from the front, back or side, and change the appearance of the generated photo (e.g. “film photo,” “mirror selfie”). There must be a bug on the mirror selfies, though, because in the feed of user-generated images, some mirrors don’t actually reflect a person — but of course, these models are not people at all. Porn Pen functions like “This Person Does Not Exist,” only it’s NSFW.
On Y Combinator’s Hacker News forum, a user purporting to be the creator describes Porn Pen as an “experiment” using cutting-edge text-to-image models. “I explicitly removed the ability to specify custom text to avoid harmful imagery from being generated,” they wrote. “New tags will be added once the prompt-engineering algorithm is fine-tuned further.” The creator did not respond to TechCrunch’s request for comment.
But Porn Pen raises a host of ethical questions, like biases in image-generating systems and the sources of the data from which they arose. Beyond the technical implications, one wonders whether new tech to create customized porn — assuming it catches on — could hurt adult content creators who make a living doing the same.
“I think it’s somewhat inevitable that this would come to exist when [OpenAI’s] DALL-E did,” Os Keyes, a PhD candidate at Seattle University, told TechCrunch via email. “But it’s still depressing how both the options and defaults replicate a very heteronormative and male gaze.”
Ashley, a sex worker and peer organizer who works on cases involving content moderation, thinks that the content generated by Porn Pen isn’t a threat to sex workers in its current state.
“There is endless media out there,” said Ashley, who did not want her last name to be published for fear of being harassed for their job. “But people differentiate themselves not by just making the best media, but also by being an accessible, interesting person. It’s going to be a long time before AI can replace that.”
On existing monetizable porn sites like OnlyFans and ManyVids, adult creators must verify their age and identity so that the company knows they are consenting adults. AI-generated porn models can’t do this, of course, because they aren’t real.
Ashley worries, though, that if porn sites crack down on AI porn, it might lead to harsher restrictions for sex workers, who are already facing increased regulation from legislation like SESTA/FOSTA. Congress introduced the Safe Sex Workers Study Act in 2019 to examine the affects of this legislation, which makes online sex work more difficult. This study found that “community organizations [had] reported increased homelessness of sex workers” after losing the “economic stability provided by access to online platforms.”
“SESTA was sold as fighting child sex trafficking, but it created a new criminal law about prostitution that had nothing about age,” Ashley said.
Currently, few laws around the world pertain to deepfaked porn. In the U.S., only Virginia and California have regulations restricting certain uses of faked and deepfaked pornographic media.
Systems such as Stable Diffusion “learn” to generate images from text by example. Fed billions of pictures labeled with annotations that indicate their content — for example, a picture of a dog labeled “Dachshund, wide-angle lens” — the systems learn that specific words and phrases refer to specific art styles, aesthetics, locations and so on.
This works relatively well in practice. A prompt like “a bird painting in the style of Van Gogh” will predictably yield a Van Gogh-esque image depicting a bird. But it gets trickier when the prompts are vaguer, refer to stereotypes or deal with subject matter with which the systems aren’t familiar.
For example, Porn Pen sometimes generates images without a person at all — presumably a failure of the system to understand the prompt. Other times, as alluded to earlier, it shows physically improbable models, typically with extra limbs, nipples in unusual places and contorted flesh.
“By definition [these systems are] going to represent those whose bodies are accepted and valued in mainstream society,” Keyes said, noting that Porn Pen only has categories for cisnormative people. “It’s not surprising to me that you’d end up with a disproportionately high number of women, for example.”
While Stable Diffusion, one of the systems likely underpinning Porn Pen, has relatively few “NSFW” images in its training dataset, early experiments from Redditors and 4chan users show that it’s quite competent at generating pornographic deepfakes of celebrities (Porn Pen — perhaps not coincidentally — has a “celebrity” option). And because it’s open source, there’d be nothing to prevent Porn Pen’s creator from fine-tuning the system on additional nude images.
“It’s definitely not great to generate [porn] of an existing person,” Ashley said. “It can be used to harass them.”
Deepfake porn is often created to threaten and harass people. These images are almost always developed without the subject’s consent out of malicious intent. In 2019, the research company Sensity AI found that 96% of deepfake videos online were non-consensual porn.
Mike Cook, an AI researcher who’s a part of the Knives and Paintbrushes collective, says that there’s a possibility the dataset includes people who’ve not consented to their image being used for training in this way, including sex workers.
“Many of [the people in the nudes in the training data] may derive their income from producing pornography or pornography-adjacent content,” Cook said. “Just like fine artists, musicians or journalists, the works these people have produced are being used to create systems that also undercut their ability to earn a living in the future.”
In theory, a porn actor could use copyright protections, defamation and potentially even human rights laws to fight the creator of a deepfaked image. But as a piece in MIT Technology Review notes, gathering evidence in support of the legal argument can prove to be a massive challenge.
When more primitive AI tools popularized deepfaked porn several years ago, a Wired investigation found that nonconsensual deepfake videos were racking up millions of views on mainstream porn sites like Pornhub. Other deepfaked works found a home on sites akin to Porn Pen — according to Sensity data, the top four deepfake porn websites received more than 134 million views in 2018.
“AI image synthesis is now a widespread and accessible technology, and I don’t think anyone is really prepared for the implications of this ubiquity,” Cook continued. “In my opinion, we have rushed very, very far into the unknown in the last few years with little regard for the impact of this technology.”
To Cook’s point, one of the most popular sites for AI-generated porn expanded late last year through partner agreements, referrals and an API, allowing the service — which hosts hundreds of nonconsensual deepfakes — to survive bans on its payments infrastructure. And in 2020, researchers discovered a Telegram bot that generated abusive deepfake images of more than 100,000 women, including underage girls.
“I think we’ll see a lot more people testing the limits of both the technology and society’s boundaries in the coming decade,” Cook said. “We must accept some responsibility for this and work to educate people about the ramifications of what they are doing.”
AI is getting better at generating porn. We might not be prepared for the consequences.