16 October, 2024
YouTube is experimenting with Notes, a crowdsourced feature that lets users add context to videos
YouTube is introducing a new experimental feature that will allow viewers to add “Notes” to provide more context and information under videos, the company told TechCrunch exclusively. If the feature sounds similar, it’s because it follows the same concept as Community Notes on X (formerly Twitter).
The Google-owned company says the feature can be used for things like clarifying when a song is meant to be a parody, or letting viewers know when older footage is being portrayed as a current event.
The launch comes during a pivotal U.S. election year. While misinformation was a significant issue during the 2020 presidential election, misinformation during the 2024 election is poised to be even more of a problem thanks to the rise of generative AI. With this new feature, YouTube is likely looking to minimize the spread of misinformation on its platform.
To start, the pilot will be available on mobile in the U.S. in English.
YouTube acknowledges that there will likely be mistakes during the test phase, as it anticipates that there may be notes that aren’t a great match for a video or that incorrect information may be included in a note. The company plans to learn from the test phase and is also welcoming feedback from viewers and creators on the quality of notes.
As part of the test phase, a limited number of users will be invited to write notes. Eligible users will have an active YouTube channel in good standing.
Viewers in the U.S. will start to see notes on videos in the coming weeks and months. During the initial pilot, third-party evaluators will rate the helpfulness of and accuracy of notes. YouTube will use these responses to train its systems.
If the third-party evaluators find notes to be helpful, they will appear under a video. Viewers will then be asked if they think the note is “helpful,” “somewhat helpful” or “unhelpful.” They’ll be asked why they think a note is helpful or unhelpful. For instance, a person can say they found a note to be helpful because it cited high-quality sources or because it was written in a neutral way.
YouTube will then use an algorithm to determine which notes are published based on the ratings they have received. The algorithm will identify notes that are helpful to a broad audience. For instance, if a significant number of people who previously rated notes differently now rate a particular note as helpful, YouTube will be more likely to display that note under a video.
The company says the system will continuously improve as more notes are submitted and rated across different types of videos on the platform. As it improves the feature, YouTube will also determine whether it makes sense to officially roll it out.
Apple kills Pay Later feature ahead of Affirm integration
Two years after it was announced at WWDC, Apple’s U.S.-only Pay Later feature is no more. TechCrunch has confirmed the news, which was first noted by 9 to 5 Mac.
Here’s the full statement offered to TechCrunch:
Starting later this year, users across the globe will be able to access installment loans offered through credit and debit cards, as well as lenders, when checking out with Apple Pay. With the introduction of this new global installment loan offering, we will no longer offer Apple Pay Later in the U.S.
Our focus continues to be on providing our users with access to easy, secure and private payment options with Apple Pay, and this solution will enable us to bring flexible payments to more users, in more places across the globe, in collaboration with Apple Pay enabled banks and lenders.
Pay Later’s life was extraordinarily short, having officially launched in late March 2023. Ultimately, however, the move isn’t surprising. Apple announced at last week’s WWDC 2024 that users would be able to access loans through third-party app Affirm through Apple Pay.
“The ability to access installments from credit and debit cards with Apple Pay will roll out starting in Australia with ANZ; in Spain with CaixaBank; in the U.K. with HSBC and Monzo; and in the U.S. with Citi, Synchrony, and issuers with Fiserv,” Apple noted last week. “Users in the U.S. will also be able to apply for loans directly through Affirm when they check out with Apple Pay.”
Affirm added of the loan feature, “This provides users with additional payment choices, and offers the ease, convenience and security of Apple Pay alongside the features users love in Affirm — flexibility, transparency and no late or hidden fees.” The feature is arriving later this year.
Starting now, meanwhile, users in the U.S. have one fewer option through Apple Pay. Pay Later is no longer offering new loans as of Monday, though those with current loans through the service will still be able to pay for and manage them through Apple Wallet.
Personal loan numbers have increased in recent years. The figure is up 8% since the outset of the pandemic.
Patreon introduces a gifting feature and other creator tools
Patreon, the paid membership platform for creators, announced Tuesday the release of new features designed to help creators monetize their non-paying followers and tap into new revenue streams. This includes the ability to sell individual posts, a gifting feature for creators to gift limited-time subscriptions to fans, and promotional tools for sharing clips on social media.
The new features reflect Patreon’s ongoing efforts to expand beyond its role as a monetization platform. Last year, Patreon launched free membership options, enabling fans to follow their favorite influencers without having to commit to a monthly subscription. The company revealed in today’s blog post that the move has already seen more than 30 million sign-ups.
Now, Patreon is introducing one-time purchases for content like posts and collections, so followers can access paywalled content without paying for a membership. The company previously allowed one-time purchases for various digital products, such as videos, podcast episodes and images.
The platform is also testing a gifting offering that lets creators give limited-time subscriptions to fans. For instance, influencers can send someone a one-month membership as a gift for being a loyal follower.
Additionally, Patreon is improving communication by opening up community chat sections to both free and paid members, as well as introducing features like a countdown clock and a live chat feature to promote new and upcoming content. There will also be a new tool for auto-generating audio and video clips of Patreon posts to use as teasers on social media platforms.
All of these new features will become available in the coming months, the company said.
Updated 6/25/24 at 12:52 p.m. ET with correction to the statement that fans can gift subscriptions to their friends. The tool only allows creators to gift memberships to fans.
Video editing app Captions releases AI edit feature that automatically adds effects to your video
Video editing app Captions, which is backed by a16z, Kleiner Perkins and Sequoia Capital, has launched a new feature that takes an existing unedited video and adds custom graphics, zooms, music, sound effects, transitions and motion backgrounds based on the content.
There are a few restrictions for AI edit — it should be a vertical video of a person talking, with only one person in the frame.
The key part is that even if you don’t have such videos or that is not your style of video shooting, you can use Captions’ AI avatars to create a video with a short prompt and later feed that video to the AI edit feature to get a fully edited video in minutes with different transitions and effects.
Gaurav Misra, co-founder and CEO of Captions, told TechCrunch that when he left Snap in 2021, he thought about making the video-making process easier with the core purpose of communication.
“Our main goal has been allowing people to convey a message. If you think about the journey of what does it take to say what you want to say on video, it’s complicated. It starts with the idea of what do I actually want to say, preparing the script, recording the video and editing it to make the clip engaging,” he said in reference to the thinking behind building a pipeline of features using AI to create videos.
Utilizing AI to create videos
Misra said that with Captions, the company wants to provide three types of video recording tools to people. First, it aims to provide the best camera toolkit to aid recording. Second, it also offers editing tools like AI-powered corrections for manually recorded video. And finally, Captions features a generative tier where a user doesn’t have to record a video at all.
The company currently offers 12 AI characters. But going forward, the company wants to add three to four characters to its portfolio every week. Eventually, the goal for the startup is to let users create their own AI characters.
Misra sees these tools being used primarily for sales, marketing and communication channels by consumer-focused companies. Companies such as D-ID and Synthesia are allowing organizations to create digital avatars for videos. Earlier this month, TikTok also allowed creators to make AI avatars along with offering its own stock AI character for ads. Misra believes that Captions offers better quality, and you can access all video creation tools on the phone.
The tension between AI and creators
The Captions app’s feature set makes it really easy for people to create videos. We created the video below just with the prompt “Dangers of AI for creators” and tapped a few buttons for creation and edits. It’s kind of jarring to hear an AI avatar talk about the dangers of AI.
These tools make it easy for people to just flood social media and the internet in general with content, whether it is good or bad. This might create a problem of discoverability for creators who are putting in hours into creating content.
Misra acknowledged that this is a valid concern but thinks that people creating with AI tools will still have to make sure that video has good content and something that people are interested in.
“Mass production of content would be possible, but to stand out, you have to have a unique message or story, which will be where people will spend more of their time now. Kinda like how digital music enabled easier music creation without playing an instrument, but instead of lowering the quality of music, it just enabled more people to become musicians and raised the creativity bar,” he said.
The company is planning to release new features for its AI avatar-based video creations, such as a skit feature where two (or two of the same) avatars can talk to each other.
Social network Butterflies AI adds a feature that turns you into an AI character
Butterflies AI, the new social network where humans and AIs interact with each other, is launching a new Clones feature that turns you into an AI character.
This latest addition builds on a feature launched to the public back in June that lets users create AI personas, called Butterflies, that automatically create posts on the social network that other AIs and humans can then interact with.
With this new feature, users can take a selfie and create themselves as a Butterfly with a developing backstory. The idea behind the feature is to give users the ability to reimagine themselves in the form of an AI character. While not everyone may be interested in creating an AI clone of themselves and having others interact with it, the new feature is aimed at people who want to have a bit of fun and visualize how their life could be different.
Butterflies AI was founded by Snap engineer Vu Tran, who came up with the idea for the startup after seeing a lack of interesting AI products for consumers outside of generative AI chatbots. With the launch of Clones, Tran says people can now insert themselves into different storylines and wonder “What if?” Users can play around with the feature to reimagine themselves as an astronaut, a pro wrestler, a celebrity or anything else. Although you can visualize these sorts of things with text-to-image generation tools, Butterflies AI goes beyond a simple photo and lets you create an AI clone that has its own personality and background that others can then interact with.
“One interesting use of clones we find people getting excited about is the idea that they can recreate and prompt a version of themselves that still represents who they are and is authentic to their real world, but gives them a buffer between needing to show exactly them or what they’re doing,” Tran said in an emailed statement. “It feels like a safer way to share yourself to the internet.”
For example, one Butterflies AI employee recently went on a trip to France and did a lot of fun activities, but isn’t the kind of person who likes taking selfies. He decided to instead create a Butterfly of himself and prompted his clone to mimic his own real trip. Although he didn’t get a photo of himself at the lavender fields in Provence, he now has an image of his AI clone there.
The launch of the new Clones feature comes a few weeks after Meta launched the ability for select creators to clone themselves to create their own custom AI characters for chatbots that other users can interact with. Butterflies AI acknowledges that the concept behind its new Clones feature is similar to Meta’s, but notes that while Meta’s is limited to select users, its own feature is rolling out widely.
Users can create a clone by going to their account profile and clicking the new Butterflies Clones button. From there, they can capture a selfie to create themselves as a Butterfly.
While some may find the idea of Butterflies AI somewhat unsettling because it represents what happens when AI starts to invade our social media feeds, it’s mainly used for entertainment purposes, and is not seen as a replacement for Instagram.
Butterflies AI closed a $4.8 million seed round led by Coatue in November 2023. The funding round included participation from SV Angel and strategic angels, many of whom are former Snap product and engineering leaders.
The app is available for free on iOS and Android.
Android's upcoming 'Collections' feature will drive users back to their apps
Google’s plan to promote Android apps outside its Play Store is starting to shape up, as a new feature called “Collections” reveals a way that users will be able to keep up with their apps from their Android home screen. At this year’s Google I/O developer conference, the company presented developers with tools to re-engage users with the apps they already have installed on their devices. However, the company at the time didn’t detail what this surface would look like, only focusing on the software development kit (SDK) that developers could access to integrate with the new offering.
Google also shared that it already has over 35 developer partners on board to test the new feature, including Spotify, Pinterest, Tumblr, TikTok and Shopify.
In a session at the conference, developers were presented with information about a new invite-only Engage SDK, designed largely to bring users back into their apps by showcasing content, promotions and deals. The surface, Google said, would “automatically organize the best and most prevalent content from apps already installed” with a focus on enabling cross-ap-continuing journeys for nearly every app category.
With a tap, users would be able to launch full-screen, immersive experiences that highlight and arrange the most important content from their installed apps.
Now, we’re starting to get an idea of what this surface will look like, thanks to an APK teardown of the Google Play Store software, performed by the blog Android Authority. Their experiments in reverse engineering reveal a still-in-development feature known as “Collections.” Previously referred to by the codename “Cubes,” the feature introduces a Play Store-powered widget for the Android home screen that organizes app updates into categories like Watch, Listen, Read, Games, Social and Shop.
When users select a given category, they would be shown all the relevant content from their apps in this space, presuming the apps had integrated with the SDK.
As Google described during the I/O session, the new on-device surface would organize what’s most relevant to users, inviting them to jump back into their apps. The company offered a few examples, as well, showing how retail apps could invite people to continue their shopping journey by highlighting what they had saved in their cart, while apps like Uber Eats might invite users to place their favorite food order. Reddit could encourage people to read and upvote interesting posts, while Spotify could feature favorite playlists.
Google said app developers could also use the feature to showcase “personalized recommendations and promotions” that would help discover other content and deals they may have missed.
The APK teardown revealed that the feature is launching first in the U.S., before rolling out to new markets, and will also be easily accessible by way of a shortcut from the Play Store app itself. That is, when you long-press on the Play Store app, there will be a link to access Collections from the pop-up menu that appears.
Google hasn’t yet responded to a request for comment on the Collections feature, but it announced earlier that the consumer experience would launch later this year. Of note, Google recently announced a new hardware and Android event, “Made by Google,” which will take place in August. We’ve heard Google expects to announce this feature more officially at that time.
Google is launching a new Android feature to drive users back into their installed apps
Figma disables its AI design feature that appeared to be ripping off Apple's Weather app
Figma CEO Dylan Field says the company will temporarily disable its “Make Design” AI feature that was said to be ripping off the designs of Apple’s own Weather app. The problem was first spotted by Andy Allen, the founder of NotBoring Software, which makes a suite of apps that includes a popular, skinnable Weather app and other utilities. He found by testing Figma’s tool that it would repeatedly reproduce Apple’s Weather app when used as a design aid.
Allen had taken to X, formerly Twitter, to accuse Figma of “heavily” training its tool on existing apps — an accusation Field now denies.
Figma AI looks rather heavily trained on existing apps.
This is a "weather app" using the new Make Designs feature and the results are basically Apple's Weather app (left). Tried three times, same results. https://t.co/Ij20OpPCer pic.twitter.com/psFTV6daVD
— Andy Allen (@asallen) July 1, 2024
The Make Design feature is available within Figma’s software and will generate UI (user interface) layouts and components from text prompts. “Just describe what you need, and the feature will provide you with a first draft,” is how the company explained it when the feature launched.
The idea was that developers could use the feature to help get their ideas down quickly to begin exploring different design directions and then arrive at a solution faster, Figma said.
The feature was introduced at Figma’s Config conference last week, where the company explained that it was not trained on Figma content, community files or app designs, Field notes in his response on X.
“In other words, the accusations around data training in this tweet are false,” he said.
But in its haste to launch new AI features to remain competitive, the quality assurance work that should accompany new additions seems to have been overlooked.
Mirroring complaints in other industries, some designers immediately argued that Figma’s AI tools, like Make Design, would wipe out jobs by bringing digital design to the mass market, while others countered that AI would simply help to eliminate a lot of the repetitive work that went into design, allowing more interesting ideas to emerge.
Or this will remove a lot of the repetitive work in design and raise the bar so new interesting ideas can emerge, just like new tooling always has
Corps that don’t invest in design still wont, and those that do will continue to do so. And the truth is there’s usually so much…
— Dustin Karp (@digitdustin) June 26, 2024
Allen’s discovery that Figma essentially seemed to be copying other apps led to increased concern among the design community.
“Just a heads up to any designers using the new Make Designs feature that you may want to thoroughly check existing apps or modify the results heavily so that you don’t unknowingly land yourself in legal trouble,” Allen warned others on X.
Field responded by clarifying that Make Design uses off-the-shelf large language models, combined with “systems we commissioned to be used by these models.” He said the problem with this approach is that the variability is too low.
(4) I have asked our team to temporarily disable the Make Design feature until we are confident we can stand behind its output. The feature will be disabled when our US based team wakes up in a few hours, and we will re-enable it when we have completed a full QA pass on the…
— Dylan Field (@zoink) July 2, 2024
“Within hours of seeing [Allen’s] tweet, we identified the issue, which was related to the underlying design systems that were created,” Field wrote on X. “Ultimately it is my fault for not insisting on a better QA process for this work and pushing our team hard to hit a deadline for Config.”
Apple was not immediately available for comment. Figma pointed to Field’s tweets as its statement on the matter.
Field says Figma will temporarily disable the Make Design feature until the team is confident it can “stand behind its output.” The feature will be disabled as of Tuesday and will not be re-enabled until Figma has completed a full QA pass on the feature’s underlying design system.
India clings to cheap feature phones as brands struggle to tap new smartphone buyers
India, the world’s second-largest smartphone market after China with around 750 million users, is struggling to get new smartphone buyers, as the average person in the country finds it hard to afford a brand-new smartphone.
Instead of switching to a new smartphone, millions of Indian consumers prefer to continue with their existing feature phones or only upgrade to a new feature phone. The latter has resulted in growth in feature phone shipments in the country despite these phones being pretty old-fashioned and having just a handful of features compared to smartphones, which help solve most on-the-go computing requirements for a typical buyer.
According to data shared by market analyst firm Counterpoint, feature phone shipments in India saw 10% year-on-year growth last year, up from -31% in 2022. This growth comes as existing feature phone buyers resist from picking a new smartphone.
Of the total feature phone market, 4G feature phone shipments in India grew to 25% in 2023 from 9% in 2022. However, 2G feature phones still dominate the Indian market, with a 75% share, per Counterpoint.
Indian telecom giant Reliance Jio leads the 4G feature phone market, with a share of 27%, followed by Transsion Holdings-owned Itel and Indian phone maker Lava at 24% and 18%, respectively, according to Counterpoint. The telecom operator offers a range of models in the segment along with dedicated tariff plans and access to services digital services, including the country’s real-time payment system Unified Payments Interface and apps such as JioCinema, JioTV and WhatsApp to attract the masses.
However, Tarun Pathak, research director for devices and ecosystems at Counterpoint, told TechCrunch that the 4G feature phones from Jio have hindered smartphone adoption in the country to a certain extent “as the consumers are now able to get basic apps on feature phones themselves.”
Jio eyeing a 5G feature phone to capitalize on the trend
Jio, the biggest telecom player in India in terms of its subscriber base and revenues, is quietly planning a 5G feature phone as brands face challenges in attracting first-time smartphone buyers in the South Asian market, TechCrunch has exclusively learned.
The telecom operator owned by the Indian conglomerate Reliance Industries has been selling feature phones for years and gradually moved to smartphones with Google in 2021. But its first smartphone was too pricey for its mediocre specifications and did not help move the needle. Now Jio sees feature phones as the strongest weapon it has to get phones into as many hands in India as possible.
Jio’s 5G feature phone will run Linux-based software KaiOS, which already powers the telco’s existing feature phone range. Contract manufacturers Dixon Technologies and NeoLync are currently prototyping the handset, sources said.
The phone would be announced at Reliance’s forthcoming annual general meeting, which is expected in August, though it is not likely to be ready for the mass market anytime soon, according to the sources. The phone’s pricing is also among the questions that Jio is yet to determine.
Sources familiar with the matter told TechCrunch that the Mumbai-based telco is in initial talks with Qualcomm for the early version of its 5G feature phone. However, other chipmakers may be involved in the final device, as the discussions are still on its development.
Qualcomm denied its participation in discussions with Jio when TechCrunch reached out this week. A Reliance Industries spokesperson declined to comment.
Alongside the new 5G feature phone, Jio plans to launch its 5G Android tablet and a 5G smartphone this year, two people familiar with the telco’s plans told TechCrunch.
Dixon Technologies and NeoLync did not respond to requests for comment.
Until now, Jio has a 4G line of devices, which includes the JioPhone Next as its first Android smartphone. The telco, however, is looking to upgrade its portfolio to 5G devices, as it aims to expand its telecom revenues and is working on monetizing 5G connectivity. Jio, along with competitors Airtel and Vi, recently hiked tariffs in the country to boost their subscription revenues. The operator is also expected to spin off from Reliance and list as a public company on Indian stock exchanges at a valuation of $112 billion as early as next year.
Jio’s feature phone journey so far
In August 2017, Jio unveiled its first 4G feature phone, JioPhone, based on KaiOS. Shortly after its launch, the JioPhone became the country’s top-selling feature phone. The candybar phone also later received support for popular apps, including Facebook, WhatsApp and YouTube to expand usage.
In 2018, Jio introduced the next-generation feature phone with a QWERTY keyboard. However, the JioPhone 2 could not repeat the success of the JioPhone due to its limited availability and appeal.
Last year, the telco introduced the Jio Bharat as a $12 feature phone series to revive its presence in the feature phone market. It also updated the original JioPhone with a face-lift called the JioPhone Prima. The launch of the new models helped grow the market of 4G feature phones in India.
Why are millions of Indians not buying new smartphones?
India’s smartphone market saw over 11% year-on-year growth in the first quarter, the third consecutive quarter of growth in shipments, with 34 million smartphones shipped during the quarter, according to IDC. However, unlike the budget price segments (between sub-$100 and $200), the super-premium segment (over $800) saw the highest 44% growth, seeing an increase in its share to 7%. It was dominated by the latest and last-generation iPhone models followed by the Galaxy S24 and Galaxy S23 families.
Nonetheless, the share of the entry-level segment (sub-$100) in the Indian smartphone market declined by 14% year-on-year to 15%, down from 20% a year ago, per IDC. It used to be 26% of the overall smartphone market in 2020 and 20% in 2021.
Market experts believe that the primary reason for the sharp decline in entry-level smartphone shipments is the reluctance of feature phone users to switch to affordable smartphones.
Navkendar Singh, associate vice president for devices research at IDC, told TechCrunch that around 350 million people in India still use a feature phone, and 50% of them have a model that’s less than $18. “Because they’re not able to buy a smartphone, they’re replacing their feature phones,” he said.
Singh also noted that India’s smartphone penetration is less than 50%, and it remained unchanged for the last couple of years, though the market saw some growth in smartphone shipments after the COVID pandemic.
“The transition from feature phones to smartphones is not easy for some consumers,” said Counterpoint’s Pathak, adding that older people and people in low-income groups and blue-collar jobs are not upgrading to smartphones in the country.
“The pricing difference between smartphones and feature phones is still wide enough to limit users to switch to smartphones easily,” he said.
According to IDC, a smartphone’s average selling price (ASP) in India is $255. It grew over 54% from $165 in 2020. The smartphone ASP in India is merely around 31% compared to the $826 in the U.S.
Still, millions of people find it hard to buy a smartphone in India due to their low disposable income.
“While India has one of the cheapest smartphones in the world, as a percentage of users’ income, it is less affordable than even in the U.S. and U.K.,” according to a report titled The State of India’s Digital Economy by the Indian Council for Research on International Economic Relations and Prosus.
Smartphone brands offer finance options to attract buyers. However, Singh told TechCrunch that the move has so far helped only existing smartphone consumers upgrade their models.
Bird Buddy’s new AI feature lets people name and identify individual birds
As hardware startups go, few are more fun than Bird Buddy. The simple innovation of mounting a connected camera on a bird feeder has developed a strong customer base and an army of knockoffs. Our own Sarah Perez is a noted fan of the gadget.
Today, the Michigan-based company announced Name That Bird, a new AI-powered feature that lets users name — and track — individual birds. While we can’t yet speak to the feature’s efficacy, it’s easy to see how it was one of the most requested among users. It’s an added kind of emotional connection to these strange little backyard visitors you watch eating on cam.
Name That Bird determines individual members of a species by identifying distinguishing characteristics that most humans would be hard-pressed to spot.
“The detail made available by the 5MP high-resolution Bird Buddy sensor present in all our feeders along with its close-up format is designed to bring out the full character of visiting birds,” CEO Franci Zidar notes in a release. “This makes Bird Buddy uniquely capable of capturing the features required to make this work. The potential is massive and I couldn’t be more excited for what this unlocks.”
Name That Bird arrives as part of a larger AI-powered feature set from Bird Buddy called Natural Intelligence. Among other things, it can also determine whether individuals are injured or sick, “something bird enthusiasts are keen to know about in order to help prevent disease spread,” the company writes.
It brings non-bird animal notifications, as well. It will spot bees, butterflies, lizards and raccoons, along with cats. While it doesn’t identify the individuals, homeowners may want to know when felines are prowling around the bird feeder for their own snack.
The new features aren’t free, however. Even smart bird feeders are getting into the subscription business. Natural Intelligence is available as part of the $6/month Bird Buddy Pro subscription.