As it continues to chase TikTok, Meta today announced a series of updates and new features for its Reels products across both Facebook and Instagram. Most notably, it’s offering a Sound Sync feature on Facebook Reels and adding support for longer Instagram Reels of up to 90 seconds, instead of 60 seconds, previously. The expansion follows TikTok’s move into YouTube territory with videos that can now be 10 minutes, instead of just 3 as before.

The company is also rolling out several more creative tools, audio tools, templates, and other options to make Reels more engaging, as TikTok’s influence — and money-making potential — continues to grow.

For starters, Meta is now leveraging existing features designed for Instagram Stories and bringing them to Reels. Among these are the Poll, Quiz, and Emoji slider stickers commonly found on Instagram Stories. These will become available as Reels features, as well.

While it may make sense to repurpose some of its most popular creative tools for Reels, in doing so, Meta runs the risk of blurring the line between its various products to the point where Reels doesn’t feel all that differentiated from Stories — instead, it could end up seeming more like an extension.

Another new Instagram Reels feature is the ability to import your own audio — an option that would allow users to add commentary or a background sound from any video at least 5 seconds long that’s stored on their Camera Roll.

Plus, Instagram is introducing templates, a new tool that will allow users to create new reels using the same structure as one they just watched. This was spotted in testing back in April and is somewhat similar to TikTok’s own templating option, which is a simpler way to get started with video creation. With Reels Templates, users are able to pull in both the audio and clip sequence, however. This helps users to recreate more complex formats just by adding their own content and trimming their clips.

Image Credits: Meta

Separately, Facebook Reels is gaining a number of new features, as well.

One key addition here is the launch of Sound Sync, a tool that will automatically sync users’ video clips to a favorite track. Syncing video to sounds has been a defining characteristic of TikTok’s short-form video platform, and a technology the company has expanded on over the years with further music-related creative effects, like those including visualizations, animations, and interactivity.

Facebook Reels users will also be able to add voiceovers to their videos, which is already common on TikTok.

A few other features are aimed at appealing to more professional creators, like desktop-based video clipping tools that will make it easier for creators who publish long-form, recorded, or live videos to test out other formats, like Reels. They’ll be able to turn their content into Facebook Reels using Creator Studio, for example. Meanwhile, gaming video creators will gain tools that will let them generate short-form, vertical Reels directly from their Live content, with dual views for both gameplay and the creator cam.

Creators will also now be able to create, edit and schedule their Facebook Reels from the desktop, in addition to iOS and Android.

Image Credits: Meta

Facebook will also push suggested reels in its Feed globally, as well as in Watch and Groups, in an effort to ramp up discovery and help creators obtain broader reach. It reminded creators, too, that Reels are prioritized and ranked based on whether or not they’re featuring original content.

While these changes may help to expand the variety of reels that are produced by creators, what Meta is arguably lacking when it comes to Reels is quality.

Feeds are often dominated by boring or dated trends — like recordings of people trying out AR or beauty filters, or videos synced audio clips that feature random or uninteresting videos uploaded from users’ phones. Pets and uninspired travel video clips are also common on Reels, as is engagement bait, where the sole purpose of the video is to drive users to the comments. Some Reels creators simply pose for the camera but have nothing to say — a tactic that may have helped them build a following when Instagram was a photo-centric app that rewarded the attractive, but is less of a draw in the TikTok era, where users are demanding authenticity, not perfection.

Despite these issues, creators can still — for now, at least — rely on Facebook and Instagram’s massive reach of billions to garner attention and engagement. But the overall feel of browsing Reels remains quite different from TikTok. On the latter, commentary, insights, vlogs, and serendipitous, but highly personalized discovery continues to reign.

To address this larger issue, Meta has been paying creators directly to make Reels through its creator fund and recently announced that it will pay out bonuses to creators for original content, too. But some are beginning to suspect TikTok’s rise combined with Meta’s hampered ability to acquire competitors under increased regulatory oversight will put an end to Meta’s time at the top as the leading social platform.

If you’re working for an advertising agency or you’re in charge of a brand, editing and exporting a video ad that works across all social and web platforms can be a lengthy and time consuming task. French startup Aive speeds up the process quite drastically as it can automatically generate all formats and all durations from one source video.

When you create a new video to showcase your brand, you usually start with a 16:9 video with a duration of a couple of minutes. It can be a good first step to create shorter versions for TV spots for instance.

But the advertising industry has changed — 30-second TV spots are no longer as relevant as they used to be. Many brands now want to advertise across different platforms, which means that you need to provide a video that integrates natively with these platforms.

For instance, on Snapchat, brands need a vertical 9:16 video if they want to insert ads in between user stories. If a company wants to create a video for the Instagram newsfeed, a square 1:1 video will work better.

Each platform has different requirements when it comes to format, duration and even content emphasis. On YouTube, you can create very short videos that last just a few seconds or longer videos that can be skipped after a few seconds. In that case, you need to display your brand’s logo as quickly as possible to make sure that people don’t skip too quickly or get the message as quickly as possible.

Image Credits: Aive

When Aive users upload a video, they get an overall creative score and a breakdown of all the strengths and weaknesses in the video. For instance, the platform tells you if the video will work well on mobile, if the brand is showcased appropriately and more.

After that, users can create automated tasks to repackage that video. You can select multiple lengths and formats at once for your output videos. Once the service has finished processing your videos, you can see the result and manually adjust some details that don’t work as well as expected.

For instance, if you think the algorithm deleted an important scene, you can replace a scene in the output video with a deleted one. Once this is done, you can download the output videos to your computer and use it across several platforms.

The startup recently raised a $3.2 million (€3 million) seed round from dozens of business angels, including Pauline Duval, Renaud Visage, Kevin Polizzi, Jean-Paul Brunier, Aurélie Jean, Julien Chaumond, Jérémie Rosseli, etc.

With this funding round, the company plans to double the size of its team from 15 to 30 people. If you add previous funding and public funding from Région Occitanie, Bpifrance and La French Tech, Aive has scored $7.5 million (€7 million) since its inception.

Because it is a software-as-a-service product, Aive also works well as a team. Users can leave comments and annotate videos for other users. There is also a built-in user rights management system, which can be useful for freelancers, clients and agencies.

While the most expensive ad agencies can already generate variations of the same video for their clients, Aive makes this process more accessible to different teams and companies. It means that you don’t have to restrict yourself to a couple of ad campaigns per year as it’s easier to refresh your ads more frequently.

It’s still early days for the startup as Aive has been testing its product in private beta with a handful of companies, such as Club Med, L’Oréal, Monoprix, Nissan and Swile. So it’s going to be interesting to see how the product evolves in the coming years.

Image Credits: Aive

TikTok is already one of the world’s fastest-growing social media platforms, where it’s even overtaken YouTube on watch time in select markets and sports over 1 billion monthly active users. Now, the company is looking to expand the range of activities its users can do when they tire of flipping through short videos. A report from Reuters today indicates the company is further investing in HTML5 games, and has already begun tests. But we understand there may be more to TikTok’s gaming strategy than web-based gaming. It appears TikTok is also working on a LIVE mobile gaming feature that would allow creators to better engage fans while live streaming.

Reuters said TikTok planned to draw on parent company ByteDance’s suite of games, beginning with minigames that have simple mechanisms and a shorter playing time. It claimed tests had begun in Vietnam. But TikTok told TechCrunch that was inaccurate, saying that Vietnam gaming testing is not something it’s currently doing.

The social video app’s move into HTML5 gaming was already known, as TikTok had previously announced its plans alongside its Zynga partnership last year. The two companies had teamed up to launch the HTML5 game “Disco Loco 3D” exclusively on TikTok. And at the time, TikTok confirmed it was already in discussions with other game makers for similar deals, telegraphing a larger gaming expansion was still to come. TikTok says it doesn’t have any new partners to name on this front as of now.

Image Credits: Zynga

Ahead of the Zynga deal, TikTok had also already launched its own game, “Garden of Good,” built in partnership with nonprofit Feeding America and focused on charitable fundraising. This one felt more like an experiment to see if TikTok users would play an in-app game.

While still a smaller effort today, gaming could grow to become a significant monetization tool for TikTok, if the games end up being ad-supported or later add paid components, like in-app purchases.

In addition to its previously announced efforts to delve into HTML5 games, TechCrunch has also learned TikTok is looking to explore LIVE games in a separate effort.

According to an investigation by mobile product intelligence firm Watchful, based in Tel Aviv, TikTok is working to add minigames to LIVE videos in its app to enhance the live-streaming experience between creators and their fans.

One LIVE game Watchful uncovered is called “Draw & Guess,” which is designed specifically to encourage interaction between creators and viewers. In this Pictionary-like game, players are given words that they then draw on the screen, and viewers try to guess what they’re drawing. The correct guesses are shown on the screen.

Watchful also noted the mobile gaming feature may allow screen-sharing, so creators could share their screen with viewers in real-time, including their camera, audio, notifications, and other alerts. Their games could be shown in either landscape mode or portrait mode, the firm said.

While today, a handful of reverse engineers publicly share details of unannounced, still-in-development features they find in popular consumer apps, Watchful has productized this type of investigation and enhanced it with technology. The company uses a combination of computer vision and flow analysis to identify and emulate app changes. Its differential analysis engine compares application versions using real user data and proprietary computer vision algorithms, it says. The studies are also backed by the firm’s mobile device labs deployed around the world.

Reached for comment, TikTok did not have anything more to share about the LIVE gaming efforts, but told us it was a totally unrelated development from the HTML5 games Reuters had described.

Games aren’t the only way TikTok is looking to enhance the TikTok LIVE Platform.

Watchful also flagged another in-development feature that feels like a cross between a game and a virtual gifting experience. It allows users to add a “Treasure Box” to LIVE videos which doles out coins after a timer is up to a random set of users.

Image Credits: TikTok app image via Watchful

And it found TikTok was working on the in-app shopping experience, TikTok Shop, which is offered today in Indonesia, Malaysia, Thailand, Vietnam, Mainland China, Hong Kong, and the U.K. Here, TikTok appears to be looking at adding a shopping bag icon to LIVE shopping videos which, when clicked, opens up a window where viewers can browse through available products with the option to click to add them to a cart for immediate purchase.

Image Credits: TikTok app image via Watchful

TikTok declined to comment on these other LIVE platform plans.

YouTube today is launching a new feature designed to help users identify the most popular parts of a video they’re watching, along with other changes for its video player. The company says it’s adding a graph that people can use to easily locate and watch the most replayed parts of a video — something that could be particularly helpful for longer videos or those that haven’t broken down their various sections using either timestamps or video chapters.

The feature was previously available as an experiment that YouTube Premium subscribers could access from YouTube.com/New, but it will now become available to all YouTube users, both free and Premium, across desktop, iOS and Android platforms.

While creators will often segment their longer videos into chapters or link to select parts via timestamps, the new graph aims to make it possible for users to scrub through any YouTube video to quickly find the most interesting parts. These segments may or not always correlate to where chapters or timestamps would be located, either — particularly in the case of certain video segments going viral, for example.

In addition, the feature could help point users more to the useful parts of tutorial videos, or to the segment that has the most action — like the most watched bit of footage from a gamer’s video.

YouTube viewers will be able to now see this subtle gray graph as they scrub through the video using the red playback progress bar. If the graph is high, then that part of the video has been replayed often, says YouTube. The video thumbnail that appears will also indicate when you’ve hit the segment that is the “Most replayed.”

Image Credits: YouTube

Alongside this launch, YouTube is also rolling out support for video chapters, the segmenting tool first launched in May 2020, to smart TVs and gaming consoles. Chapters allow users to jump forward to a particular section of a video or easily rewatch that section. Since launching, YouTube has made it possible for chapters to be automatically generated and notes they’ve now been added to some 20 million videos to date.

Another newly added feature called Single Loop will now allow users to put an individual video on endless repeat. To access this feature, you’ll select the option from the same menu where you can adjust the video quality, turn on captions and more.

Also, in full-screen mode, a new panel will now appear to display information about the video — like the video description, video chapters and comments. Users will be able to access quick action buttons here to like, dislike, comment, share or add a video to a playlist directly from full-screen mode.

Image Credits: YouTube

Finally, YouTube teased it will “soon” begin another new experiment that will allow users to seek the exact moment in a video that they want to watch. This, too, will first be made available to Premium subscribers via the YouTube.com/New website, it says.

The ultimate achievement to some in the AI industry is creating a system with artificial general intelligence (AGI), or the ability to understand and learn any task that a human can. Long relegated to the domain of science fiction, it’s been suggested that AGI would bring about systems with the ability to reason, plan, learn, represent knowledge, and communicate in natural language.

Not every expert is convinced that AGI is a realistic goal — or even possible. But it could be argued that DeepMind, the Alphabet-backed research lab, took a toward it this week with the release of an AI system called Gato,

Gato is what DeepMind describes as a “general-purpose” system, a system that can be taught to perform many different types of tasks. Researchers at DeepMind trained Gato to complete 604, to be exact, including captioning images, engaging in dialogue, stacking blocks with a real robot arm, and playing Atari games.

Jack Hessel, a research scientist at the Allen Institute for AI, points out that a single AI system that can solve many tasks isn’t new. For example, Google recently began using a system in Google Search called multitask unified model, or MUM, which can handle text, images, and videos to perform tasks from finding interlingual variations in the spelling of a word to relating a search query to an image. But what is potentially newer, here, Hessel says, is the diversity of the tasks that are tackled and the training method.

DeepMind’s Gato architecture.

“We’ve seen evidence previously that single models can handle surprisingly diverse sets of inputs,” Hessel told TechCrunch via email. “In my view, the core question when it comes to multitask learning … is whether the tasks complement each other or not. You could envision a more boring case if the model implicitly separates the tasks before solving them, e.g., ‘If I detect task A as an input, I will use subnetwork A. If I instead detect task B, I will use different subnetwork B.’ For that null hypothesis, similar performance could be attained by training A and B separately, which is underwhelming. In contrast, if training A and B jointly leads to improvements for either (or both!), then things are more exciting.”

Like all AI systems, Gato learned by example, ingesting billions of words, images from real-world and simulated environments, button presses, joint torques, and more in the form of tokens. These tokens served to represent data in a way Gato could understand, enabling the system to — for example — tease out the mechanics of Breakout, or which combination of words in a sentence might make grammatical sense.

Gato doesn’t necessarily do these tasks well. For example, when chatting with a person, the system often responds with a superficial or factually incorrect reply (e.g., “Marseille” in response to “What is the capital of France?”). In captioning pictures, Gato misgenders people. And the system correctly stacks blocks using a real-world robot only 60% of the time.

But on 450 of the 604 aforementioned tasks, DeepMind claims that Gato performs better than an expert more than half the time.

“If you’re of the mind that we need general [systems], which is a lot of folks in the AI and machine learning area, then [Gato is] a big deal,” Matthew Guzdial, an assistant professor of computing science at the University of Alberta, told TechCrunch via email. “I think people saying it’s a major step towards AGI are overhyping it somewhat, as we’re still not at human intelligence and likely not to get there soon (in my opinion). I’m personally more in the camp of many small models [and systems] being more useful, but there’s definitely benefits to these general models in terms of their performance on tasks outside their training data.”

Curiously, from an architectural standpoint, Gato isn’t dramatically different from many of the AI systems in production today. It shares characteristics in common with OpenAI’s GPT-3 in the sense that it’s a “Transformer.” Dating back to 2017, the Transformer has become the architecture of choice for complex reasoning tasks, demonstrating an aptitude for summarizing documents, generating music, classifying objects in images, and analyzing protein sequences.

The various tasks that Gato learned to complete.

Perhaps even more remarkably, Gato is orders of magnitude smaller than single-task systems including GPT-3 in terms of the parameter count. Parameters are the parts of the system learned from training data and essentially define the skill of the system on a problem, such as generating text. Gato has just 1.2 billion, while GPT-3 has more than 170 billion.

DeepMind researchers kept Gato purposefully small so the system could control a robot arm in real time. But they hypothesize that — if scaled up — Gato could tackle any “task, behavior, and embodiment of interest.”

Assuming this turns out to be the case, several other hurdles would have to be overcome to make Gato superior in specific tasks to cutting-edge single-task systems, like Gato’s inability to learn continuously. Like most Transformer-based systems, Gato’s knowledge of the world is grounded in the training data and remains static. If you ask Gato a date-sensitive question, like the current president of the U.S. chances are it would respond incorrectly.

The Transformer — and Gato, by extension — has another limitation in its context window, or the amount of information the system can “remember” in the context of a given task. Even the best Transformer-based language models can’t write a lengthy essay, much less a book, without failing to remember key details and thus losing track of the plot. The forgetting happens in any task, whether writing or controlling a robot, which is why some experts have called it the “Achilles’ heel” of machine learning.

“It’s not that Gato makes new things possible,” Guzdial added, pointing to the system’s shortcomings. “[B]ut it makes clear we can do more with modern machine learning models than we thought.”

Pinterest on Monday launched a new app aimed at making it easier for creators to livestream to its platform. The new Pinterest TV Studio app for iOS and Android will allow select creators to go live on Pinterest as well as use multiple devices in order to achieve different camera angles, the app’s description states.

The company didn’t announce the app’s debut — perhaps because it’s not broadly available to all creators at this time. Instead, upon first launch, creators have to enter in a code or scan a barcode that Pinterest provides in order to gain access to the livestreaming tools the app provides.

Image Credits: Pinterest

However, the addition of a dedicated livestreaming app is another example of how Pinterest is rethinking its place in the broader social media landscape, where TikTok’s rise has pushed other platforms to adopt a video-first focus. Today, major social players including Instagram, Facebook, Snapchat and YouTube have a TikTok-like short video feature available, and most offer features for creating live video content.

Pinterest, meanwhile, has been working on its own video efforts with the launch of Idea Pins, a sort of video-first mashup of both TikTok-stye short-form video content and tappable Stories. The company during its earnings last week said it’s seen over 25% growth in the save rate of Idea Pins quarter over quarter. And Pinterest users who follow multiple video creators on the site tend to visit Pinterest more often than those who do not, the company shared.

Meanwhile, Pinterest last November also announced the launch of Pinterest TV — a series of live, shoppable videos from creators focused on food, home, fashion, beauty, DIY and more. The shows air live, then later become available for on-demand viewing. The platform was initially only available to select, hand-picked creators at the time of launch, including folks like Christian Siriano, Monica Suriyage, Tom Daley, Manny MUA, Robyn Schall and others. It had earlier in the year tested livestreamed events inside the Pinterest app with top creators.

According to data from Sensor Tower, the brand-new Pinterest TV Studio app has been live on the App Store and Google Play since May 2, 2022. It has not seen enough downloads to rank the app on any app store charts and the firm isn’t able to estimate the total downloads at this time.

However, the app is available in several markets outside the U.S., including Canada, Australia, the U.K. and Germany — signaling a potential global expansion of Pinterest TV efforts.

Pinterest confirmed the launch with a statement but said it didn’t have more to share about the app at this time.

“With more Creators developing innovative programming with Pinterest TV on the Platform, we’re continuously experimenting with new ways to help Creators bring their ideas to life,” a spokesperson said.

TikTok is introducing a new way to lure advertisers to its platform by giving them the ability to showcase their brands’ content next to the best videos on TikTok. Ahead of its NewFronts presentation to advertisers scheduled for this afternoon, TikTok announced the launch of TikTok Pulse, a new contextual advertising solution that ensures brands’ ads are placed next to the top 4% of all videos on TikTok. Notably, the solution will also be the first ad product that involves a revenue share with creators,

The company said creators and publishers with at least 100,000 followers on TikTok will be eligible for the revenue share program during the initial stage of the TikTok Pulse program.

TikTok told TechCrunch the Pulse program will launch to U.S. advertisers in June 2022 with additional markets to follow in the fall.

TikTok didn’t say how many creators it would actually approve for the program in the early phases. But longer-term, the move could help TikTok to attract more creators to its social video app, following its earlier investments in creator monetization. Last December, TikTok debuted an online “Creator Next” portal where it organized all the tools creators can use to make money on its app in one place. Here, creators can learn about how to accept virtual gifts and payments from fans viewing their videos and their TikTok LIVE content. They can also apply to the Creator Marketplace to be connected with brands for sponsored content if they have at least 10,000 followers.

Now, TikTok will be able to add advertising revenue share to that list of creator monetization opportunities.

Image Credits: TikTok

The new program isn’t just about helping creators, however. It’s also about ensuring advertisers a more “brand-safe” environment for their content, similar to something like YouTube’s Partner Program (YPP).

On YouTube, the YouTube Partner Program allows creators to earn ad revenue from display, overlay, and video ads on their channel, in addition to providing access to other features like channel memberships, Super Chat, a merch shelf, and more. For advertisers, meanwhile, YPP allows them to reach videos from creators with more traction and subscribers, whose channels have also been vetted by YouTube for adhering to its content policies. This helps brands control their ads’ placement, so they don’t accidentally end up posted alongside hate speech or misinformation, for example.

TikTok says its new TikTok Pulse program will also be focused on making sure the creator content is “suitable” for advertisements.

“Our proprietary inventory filter ensures that TikTok Pulse ads are running adjacent to verified content with our highest level of brand suitability applied on the platform,” an announcement from TikTok states. “Additional post-campaign measurement tools such as third-party brand suitability and viewability verification provide advertisers the opportunity and transparency to analyze and understand the impact of their campaigns,” it noted.

The company says the program will additionally offer brands a way to target their ads to particular areas of TikTok. Through Pulse, brands can place their ads alongside videos across 12 categories of content, including things like beauty, fashion, cooking, gaming, and more.

At launch, only advertisers TikTok invited to join the program will have access to TikTok Pulse. But a TikTok spokesperson said the plan is to roll out the solution to more brands in the months that follow.

TikTok declined to share other specific details about the new program, like the revenue share percentage for creators, ad pricing, or information about how soon someone browsing their For You page would see Pulse ads appear, among other things.

Later this afternoon, TikTok is scheduled to pitch its platform to advertisers at the NewFronts, where TikTok’s GM for its North America Global Business Solutions, Sandie Hawkins, and its Global Head of Business Marketing, Sofia Hernandez, will talk about TikTok Pulse as well as explain to advertisers why TikTok should be a part of their media buy considerations.

YouTube today confirmed the site is experiencing a variety of issues that are impacting users worldwide. The company acknowledged the service’s problems after numerous users reported error messages when trying to watch videos or noted various site elements weren’t available — like the sidebar navigation or the Setting menus, for example — among other things.

Some users also found they couldn’t sign in to their YouTube account or switch between profiles. Others said they were unable to cast YouTube to their TV or use the app on their gaming console.

When trying to play videos, some would see an error that said “No Internet Connection” even though they weren’t otherwise having connectivity issues.

In a tweet, YouTube said it’s aware of the problems and is working on a fix. The company also noted the issues were not limited to a geographic region but were global in nature and taking place “across devices.”

The web page YouTube linked to from its tweet includes a list of the most critical problems that are affecting users, but it’s not a complete list.

On Twitter, the company also acknowledged the problems could be causing problems with users being unable to navigate YouTube Studio or stream from its live TV service, YouTube TV. Users trying to stream YouTube TV via Chromecast and Google TV said they were being logged out and were unable to log back in, for instance.

YouTube’s Twitter account also responded to users who reported their subscriptions and videos had disappeared.

So far, YouTube is not entirely down and users aren’t experiencing the same set of problems. But if you’re having any issues with YouTube or its apps today, this would likely be why.

YouTube has not yet explained what’s causing the problem with its services, but these sorts of semi-outages often have to do with the underlying technology, like the cloud services.

The company said further updates, as available, will be published to its support site here: https://support.google.com/youtube/thread/159624032. The post was published 2 hours ago, but the issues were being reported by users long before that.

YouTube has not provided an ETA to a fix.

Last August, TechCrunch noted TikTok was working to build its own augmented reality (AR) development platform called Effect House, which would allow creators to build AR effects for use in TikTok’s video app. Today, the Effect House has officially gone live. The platform has been operating in a close beta since last fall, where it’s so far been adopted by over 450 creators who have gone on to publish their effects on TikTok, where they’ve been used in 1.5 billion videos gaining over 600 billion views globally, says TikTok.

The company is now opening up beta access to encourage further development.

The launch will put TikTok in closer competition with both Snap and Meta, both of which already offer a range of tools that allow developers to build out AR experiences and effects for their own respective families of applications. Meanwhile, TikTok’s vast library of effects is a key part of what powers its video service today, and help inspire creators to make videos. By allowing creators to build their own effects, TikTok will be able to prompt even more video creation.

While building tools with Effect House may require some technical know-how, the company has created a detailed set of documentation that teaches users how to use the platform to create specific types of effects, like Segmentation, Face Mask, Head Tracker, Face Stretch, 3D Face as well as how to use different textures, materials, lighting and shadows, and more. The site also features templates, online tutorials, and a Knowledge Lab where TikTok engineers will live demo how to create a particular effect using Effect House. One demo already available shows creators how TikTok made the TikTok “Bonk!” effect and shares tips about how to make similar effects, for example.

Image Credits: TikTok

TikTok also published a set of dedicated guidelines for the Effect House, which detail the additional policies beyond the existing Community Guidelines that effect creators must adhere to. These explain that TikTok won’t allow creators to publish effects that promote colorism, negative stereotypes against protective groups, those that depict cosmetic surgery (e.g lip fillers), or those that encourage scrutiny of someone else’s appearance.

The latter is an interesting choice on TikTok’s part, as some of the most popular filters and effects to date across platforms are the “beauty filters,” that smooth skin, adjust jawlines, change the shape of facial features, apply makeup, plump up the lips and more. TikTok has its own set of these, as well. But there’s been a growing backlash about this sort of editing, as the impacts of filters on users‘ mental health — and particularly on younger girls — have become more obvious. A majority (59%) of Americans now say they find beauty filters troubling, noted a Consumer Reports survey last fall.

Lawmakers, in particular, have taken interest in how social platforms are affecting how young people see themselves, as they eye bills to regulate the industry. They queried social media companies last year, including Snap, Instagram, YouTube and TikTok, about things like eating disorder content. It’s likely TikTok didn’t want to create an unnecessary minefield of potentially problematic creative effects, given the app — already popular with younger people — is also allowed to serve an under-13 crowd through an age-gating system.

TikTok says the user-created effects will be reviewed by its Trust and Safety team to help ensure they adhere to the company’s policies before they’re released to the public. And if the company misses a violative effect, the community can report it using the “Report Effect” feature. This will prompt the team to re-review it for adherence to the Effect Guidelines. The company also says if it finds that an effect is being used in a way that violates the guidelines (even if the effect itself was not in violation), it will remove the effect and the videos that used it.

And TikTok will hide some of the creative effects, like those depicting alcohol, from appearing in the Effect Try in the app.

The company kicked off the launch of the Effect House by promoting a number of early adopters of the tools, including:

• Balraj Bains, @blahblahbalraj: Creator of the Rainbow Hearts and Butterflies effects.
• Laura Gouillon, @lauragouillon: Creator of the Which Bread Are You and Which Milk Are You effects
• Allan Gregorio, @allangregorio: Creator of the What Art Style Am I and What’s Your Drag Name effects.

When creators’ effects built with Effect House are displayed on TikTok, the effects’ results page will feature the creator’s username and link to their profile. All the creator’s published effects in their portfolio can be viewed from a designated tab on their Profile page, as well.

Though open to the public, Effect House is still considered a beta, notes TikTok.

YouTube TV is catching up to other streaming apps including Netflix, Hulu, HBO Max, Prime Video and others by finally adding support for picture-in-picture on iOS devices running iOS 15 or higher. First introduced to iPad users with iOS 13 then to iPhone with iOS 14 in 2020, YouTube has been slow to adopt this now standard functionality which allows users to stream videos in a small, floating miniplayer on their homescreen while multitasking on their device.

To try out the new functionality with the YouTube TV streaming service, users first select a title they want to play then swipe up from the bottom of the screen to enable the picture-in-picture mode, the company explained in a tweet announcing the news on Wednesday. YouTube said the feature was “slowly rolling out,” so not all users may see the option immediately.

Users have been frustrated that YouTube has taken so long to implement this now-basic functionality. There is some speculation that this delay has to do with how the YouTube player on iOS has been designed for serving and tracking ads — a core aspect of YouTube’s business and one which some other streamers — like Netflix and Amazon —  wouldn’t have to contend with.

This seems like a probable explanation, given that YouTube didn’t simply add the option for picture-in-picture (PiP) as other video apps have done, but rather has been “testing” the feature since last August with Premium subscribers as part of its YouTube “experiments.” In 2020, YouTube had made opting into tests of upcoming features called “experiments” a  perk for its Premium subscribers. Last year, YouTube began testing PiP on the iOS YouTube app with subscribers in the U.S. Now, according to YouTube’s experiments website, those tests continue. On the site today, YouTube says the current test will be available until April 8.

Unfortunately, that test’s end date may not mean YouTube will add PiP support on iOS shortly thereafter. Pushed for some sort of sense as to when YouTube users would be able to access this feature, a spokesperson told TechCrunch the company aims to make PiP support available to all YouTube users on iOS — including both free users and Premium subscribers — in the “coming months.” Sigh. 

As competition heats up for creator talent, Pinterest today announced it was more than doubling its initial investment in its Creator Fund with an additional $1.2 million in a combination of cash grants, ad credits, and other creator resources for underrepresented groups. The comapny last year announced the debut of its $500,000 Creator Fund alongside new content policies and other creator tools. However, even with the fund’s increase and Pinterest’s other creator commitments, Pinterest’s investment remains smaller than the massive efforts from other social giants, including Meta, YouTube, TikTok, and Snap.

After announcing its Creator Fund in April 2021, Pinterest last fall said it would invest another $20 million toward Creator Rewards in the U.S., which would pay creators directly for participating in “challenges.” That effort is not considered a part of the Pinterest Creator Fund, however. The company says the fund focuses specifically on supporting creators from underrepresented groups by offering both financial and educational resources. (Pinterest would not disclose how much of the new $1.2 million is going towards cash grants when we asked.)

With the Creator Fund’s expansion, Pinterest will invest in creators across quarterly cycles that are now five weeks long and focus on different content areas, including Fashion/Beauty, Wellness, Lifestyle/Home, and Food. The first of the four cycles in 2022 will be focused on fashion and beauty, and will see Pinterest working for the first time with a brand partner, L’Oréal USA. The beauty giant will provide creator fund participants with its beauty industry insights and support of experts in the field, in addition to the training Pinterest provides. Creators are being invited to apply online here.

In addition to the cash grants, ad credits and an equipment stipend, the Creator Fund recipients will be presented with opportunities for brand partnerships, access to creator-focused conferences, and will get a first look at new Pinterest products, the company says.

Yet even with today’s sizable bump from the original fund —  and the $20 million in rewards announced earlier — Pinterest is still left with some steep competition for creator talent.

For comparison, TikTok launched its own creator fund in 2020 with a $200 million commitment but said it would grow to over $1 billion over the next three years. Meta also committed $1 billion to creator bonus programs. YouTube, meanwhile, last year announced a $100 million fund for Shorts creators, with the goal of specifically targeting TikTok, but touted a broader creator investment that had totaled over $30 billion over the past three years. And Snap more recently said it had paid out $250 million to creators on its TikTok clone, Spotlight, over the course of 2021.

Pinterest, meanwhile, is still getting started with its creator efforts. Over the past year or so, it’s been attempting to shift its platform from being an image pinboard for inspiration, ideas, and commerce to one that can serve the creator community through video and video-related monetization efforts. In May 2021, Pinterest officially introduced Idea Pins, a video-first feature aimed at creators which is something like a mix of both TikTok and Stories. The Pins allowed Pinterest users to record and edit creative content using tools similar to other social platforms, like background music, transitions and various interactive elements. But unlike TikTok videos, Pinterest’s Idea Pins offer a combination of video and other content — like pages where the creator shares their recipe or the steps to reproduce a DIY project, for instance.

Last fall, Pinterest also introduced a “Watch” tab in its app where users could more easily scroll through its video Pins. It’s unclear how well these efforts are paying off for Pinterest as of yet. The Information recently reported the company tried to buy an online shopping platform Verishop, but delayed its decision amid talks of a potential Pinterest takeover. Verishop by then had soured on the deal.

Pinterest says its Creator Fund graduates gained 60% more followers, on average, after completion of its program. But without a baseline to compare that to it’s difficult to understand how well the creators are really doing, or how that’s impacting Pinterest more broadly.

Pinterest says it plans to expand its Creator Fund outside the U.S. in the second half of 2022.

Reddit is exploring the idea of bringing more user-generated video content to its online discussion forums, the company has confirmed. Not much has yet to be determined about this potential new video feature — which hasn’t been launched into testing, we should note. But we understand it may involve the use of TikTok-like video editing tools including, most notably, the ability for people to “react” to videos posted by others by adding their own video to another’s.

The ability to combine videos from different people grew popular on TikTok through features known as Stitches and Duets, which have has since been adopted by TikTok rivals including Instagram’s Reels and Snapchat’s Spotlight, to varying degrees.

In Reddit’s case, however, the goal is not necessarily to build out its own TikTok competitor aimed at creators as other large social giants have done, but rather create a video tool that would continue Reddit’s mission of enabling its users to engage in discussions on topics they care about. Today, many of these discussions on Reddit’s site are written in text, but video reactions could add a new layer to these online conversations. (To be clear, video is already supported on Reddit’s platform, but video reactions — similar to “Stitches” — are not.)

The company will be reaching out to various communities, known as subreddits, to determine if they would be interested in exploring video in such a way. Specifically, it will be looking to particular communities where it thinks the video feature could be a good product fit. But those community tests have yet to begin.

“In line with our work to help people engage in the topics that matter to them through social audio, video, text, memes, and more, we’re in the process of reaching out to a few Reddit communities to see if a new video feature we’re working on is something they find useful and fun,” a Reddit company spokesperson confirmed to TechCrunch. “After getting feedback from Redditors, we’ll explore an initial test for this new capability,” they noted.

The potential “reactions” feature in the new video project was first uncovered by iOS developer Steve Moser. Other video editing tools could also be a part of the tests, including those for adding effects, filters, and stickers to videos backed by music, among other things. Some of this functionality was added to the Reddit camera app last November, as part of Reddit’s ongoing video efforts.

In December 2020, Reddit signaled its interest in expanding further into video after it acquired TikTok competitor Dubsmash. The company shut down the app this year after first working to integrate the startup’s video creation tools into Reddit. As part of that integration, Reddit announced it would introduce new camera features, including the ability to set a timer and change recording speeds, use effects, add voiceovers, trim and adjust multiple clips, and more. Reddit users today can access these video features by way of the “new post” tool, then selecting the video option and choosing “camera” to record a video.

In addition, Reddit last summer tested a TikTok-like video feed on its iOS app which, when tapped, would show a stream of short videos in a vertical feed. Users could upvote, downvote, comment, gift an award, or share the video from this feed, then swipe up to see more.

Neither of those projects are directly related to the forthcoming video test, which is focused on exploring a new use case for video on the platform. But the underlying video technology acquired from Dubsmash would be involved.

As this product is still in extremely early stages — not even alpha testing yet — a lot could still change. There’s also the possibility that Reddit’s communities show no interest in testing such a video product and the idea is scrapped.

The video feature appears to be inspired by TikTok, which already allows creators to engage in discussions using video comments. And this functionality has been subsequently cloned by other social apps including Pinterest, Twitter, and Instagram. But perhaps video discussions on Reddit more closely aligns with an idea proposed by an early 2000’s startup called Seesmic (clearly ahead of its time), which once theorized that people would respond to posts, blogs, and other text-based content, by leaving video comments. What’s old is new again, it seems.