TechCrunch+ roundup: Vanity metric dangers, planning for failure, Black founders survey

Few startups launch with a coherent content strategy.

In the early days, every project is a sprint, and there are times when putting on a show for investors can feel more important than actually serving your customers.

Blogs are a great example: Because they’re a cheap way to drive SEO, companies crank them out, then use KPIs like time on site, pages per session and social media likes to demonstrate how successful they’ve been.

“The truth is: vanity metrics don’t measure how engaged potential customers are,” writes Christopher P. Willis, chief marketing and pipeline officer at Acrolinx.

Full TechCrunch+ articles are only available to members
Use discount code TCPLUSROUNDUP to save 20% off a one- or two-year subscription

Relying on vanity metrics is like attending a Little League awards dinner: Everyone goes home a winner!

“They simply gauge the relative popularity of your business. This makes measuring ROI tricky.”

Creating a consistent brand strategy isn’t a major investment, and creating a shared style guide for marketing, design and sales generates positive ROI. With a content governance plan, any startup can track which offers are most likely to convert new customers.

“The biggest benefit of this is content that establishes trust,” writes Willis.

Thanks very much for reading TC+!

Walter Thompson
Editorial Manager, TechCrunch+
@yourprotagonist

Banish vanity metrics from your startup’s pitch deck

Image Credits: We Are (opens in a new window) / Getty Images

It’s legitimately nice to give your hard-working team targets they can work toward, but vanity metrics (e.g. X email signups in Y days, 20% more retweets) are like a Little League awards dinner: Everyone goes home a winner!

“The truth is, investors know what traction looks like,” writes Haje Jan Kamps, which means feel-good stats have no place in a pitch deck.

“Don’t confuse fluffy numbers and vanity metrics with your go-to-market strategy.”

3 Black founders predict little will change in VC in 2023

Image Credits: tifonimages (opens in a new window) / Getty Images

A rising tide lifts all boats, but when free-flowing venture capital starts to recede, underrepresented founders are the first to find themselves on dry ground.

Dominic-Madori Davis spoke to three Black founders to get their thoughts on the current funding landscape and the issues that are top-of-mind for them as we head into the new year:

Vernon Coleman, founder and CEO, Realtime
Sevetri Wilson, founder and CEO, Resilia
Abimbola Adebayo, founder and CEO, Pinnu Analytics

The fundraising stages are not about dollar values — they’re about risk

Image Credits: Richard Drury (opens in a new window) / Getty Images

Before a founding team approaches any investor, they’ll need a clear idea of how their planned company will make money.

And also: how it will lose money.

Investors are open to ideas, but because they view everything through a lens of risk, entrepreneurs must develop a holistic understanding of where it exists in their company.

“‘For our company to be successful, these three things have to be true’ is a potent phrase in the earliest stages of starting a company,” writes Haje Jan Kamps.

With IT spending forecast to rise in 2023, what does it mean for startups?

Image Credits: We Are / Getty Images

The fact that so many CIOs and analysts believe IT spending will increase in 2023 is potentially good news for new SaaS companies hoping to weather this downturn, but “it’s not all rosy,” writes Ron Miller.

To bring these predictions down to earth, he interviewed several investors, industry watchers and CIOs to get their thoughts on “what’s coming for enterprise startups in 2023.”

TechCrunch+ roundup: Vanity metric dangers, planning for failure, Black founders survey by Walter Thompson originally published on TechCrunch

Amazon adds free music videos, viral videos and more ad-supported content to Fire TV

Amazon has announced that Fire TV users in the U.S. can now watch tens of thousands of music videos from major and independent labels for free. No downloads, fees or subscriptions are required to watch the music videos. The ad-supported music videos will be available from artists on the Billboard Hot 100, including Taylor Swift, Drake, Harry Styles and Lizzo, as well as a catalog of classics from different genres. The company says new content will be added daily.

Fire TV customers can find personalized recommendations based on their likes and viewing history, create their own mixes or choose from pre-made playlists, such as Top Holiday Hits, Best of 2022 Recap and Country Today. Users have unlimited music video skips and also have the option to play a continuous stream of similar music videos.

You can access the music videos by pressing the voice control on the Fire TV remote and saying “Alexa, find Music Videos.” Or, you can search “Music Videos” in the Appstore then click the “Music Videos on Fire TV.” From there, you need to click “get” app to download. When the download is complete, you can select “open” to access the free music videos.

In addition to music videos, Fire TV customers will now have access to additional ad-supported content, such as business and finance news from Bloomberg, The Street, CNBC and others. Users will also get access to entertainment news from brands like E! News and Mixible. In addition, customers can watch game previews and trailers, gaming news, developer interviews, how to’s, esports and more from providers including IGN, ESTV and Crown Channel. Last, users can watch viral videos from Always Funny Videos, FailArmy, People Are Awesome and The Pet Collective.

You can access the new additional free content by navigating to the “Home” icon on the Fire TV navigation bar, or by pressing the “Home” button on the Alexa Voice remote. Then, you need to scroll down to “Business & Finance News”, “Entertainment News” or “Gaming News & Esports.”

The announcement comes a few months after Amazon added free movie trailers, lifestyle content, sports highlights and more to Fire TV. Given that more viewers are gravitating to free and ad-supported content as streaming subscription prices continue to increase, it makes sense for Amazon to add more free content to Fire TV.

Amazon adds free music videos, viral videos and more ad-supported content to Fire TV by Aisha Malik originally published on TechCrunch

Bird tanks on word that a Canadian company is saving its bacon

Scooter company Bird is certainly down, but it’s not yet out.

The five-year-old firm says that it and Bird Canada — a separate, private company that licenses Bird’s software and name — will merge “as soon as possible,” with an “estimated total of $32 million in new financing” from Bird Canada’s investors, “$4 million of which has already been funded.”

Bird’s stock price dropped more than 7% today amid the announcement. The stock was trading around $0.16 per share when this story was published.

The merger news follows a rough year for Bird, which recently exited dozens of markets, and alerted investors that it had inflated its revenue for two years and might have to shutter some of its operations.

In May 2021, Bird was reportedly valued at $2.3 billion, before it closed a SPAC deal that brought it to the New York Stock Exchange. Bird’s market cap has cratered since then, and now sits around $48 million.

Bird Canada was launched by Toronto Raptors founder John Bitove back in 2019. Bitove also founded Obelysk, one of Bird Canada’s investors.

According to Bird’s statement, the deal “would add additional profitable operations to Bird’s global platform, while consolidating North American operations.” Should the merger close, Bird intends for Bird Canada CEO Stewart Lyons to become the combined company’s president, and FreshBooks chief financial officer Michael Washinushi will take over for current Bird CFO Ben Lu.

Bird tanks on word that a Canadian company is saving its bacon by Harri Weber originally published on TechCrunch

Magic Eden exec sees NFT gaming like the ‘early days of mobile gaming’

Blockchain games have grown exponentially over the past year as a new and innovative alternative to the traditional gaming world. While the two areas have been widely separated, some market players see an integrated future.

“I was around in the very early days of mobile gaming, right after the iPhone came out, the App Store came out,” Chris Akhavan, chief gaming officer at NFT marketplace Magic Eden, said to TechCrunch. “I remember the attitude back then amongst traditional gaming companies was that mobile games were stupid.”

These gaming conglomerates viewed mobile games as “really small, unimpressive games” that people wouldn’t want to play, Akhavan said. “Largely, a lot of the big traditional gaming companies ignored mobile [games] for the first couple of years and that created opportunities for new mobile gaming companies like King, which is now owned by Activision, to create that margin and grow substantially.”

As that happened, big gaming companies became hyper-focused on the mobile gaming space and began acquiring smaller games to compete, Akhavan noted.

“We think that the same journey is going to happen in web3,” Akhavan said. “Over the last year, there’s been multiple billions of dollars invested in new web3 gaming studios and they’ll lead the charge here in creating the ecosystem and showing the true sides of the opportunities, which is going to be massive.”

Magic Eden exec sees NFT gaming like the ‘early days of mobile gaming’ by Jacquelyn Melinek originally published on TechCrunch

Remembering the startups we lost in 2022

It’s been a year. This roundup is never a particularly fun one to write. No one wants to see startups fail, but we’re all keenly aware that most ultimately do. A commonly cited figure suggest that 90% of these companies will ultimately fail. But even with that in mind, 2022 just hit different.

The previous two years were unprecedented in startup land, of course. Some startups blossomed and others struggled amid shutdowns and job losses. Then came the rise and fall of the SPAC wave and global supply issues. Now it’s the economy, stupid. According to figures from Crunchbase, Q3 venture capital dropped a mind-boggling 33% from last quarter and 53% from the same time last year.

The days of the $20 million seed round appear to be over — at least for now. It is, frankly, a bad time to be raising and, by extension, a bad time to be running an early-stage startup. Accordingly, this year saw a lot of startups pumping the brakes or pulling the plug. As such, this is by no means a comprehensive list. And with the continued spiral of the crypto firm, it seems we’re not out of the woods yet.

With all of that in mind, let’s take a look at some of the startups that didn’t make it.

Airlift

Airlift, once one of Pakistan’s most richly valued and funded startups, shut down in July due to lack of capital and an unsuccessful attempt to close a funding round. Before that, the commerce service platform raised $85 million in the country’s largest Series B funding, at a valuation of $275 million. The fall from those heights, thus, didn’t just impact employees and investors, but also general enthusiasm about the Pakistani tech ecosystem.

Argo AI

Image Credits: Argo AI

It wasn’t from lack of interest — or money. Argo AI had the support of two of the world’s largest carmakers: Volkswagen and Ford. Founded in 2016 by Google and Uber vets, the Pittsburgh-based firm managed to drum up $1 billion in funding over its half-dozen-year existence. Back in October, however, management dropped a bombshell during an all-hands: Argo was shutting down.

The technology and some employees would be absorbed into either Ford or VW, and the rest of its 2,000+ employees would be getting severance. Ultimately, it seems, the company failed to bring on new investors and drum up additional funds from existing backers. The dream of autonomous driving certainly isn’t going away any time soon, and both automakers want to get there, whether via in-house development or third-party acquisition. Unfortunately, however, Argo won’t be around to play a part.

Fast

Fast, a startup that provided online checkout products, announced in early April that it would shut down after days of chatter that its future was in doubt. Apparently, its 2021 revenue growth was modest — just six figures — and its cash burn was high, with no fundraising prospects in sight.

The company — founded by Domm Holland and Allison Barr Allen — was one of those that had plenty of hype around it, so its demise (especially after raising $124.5 million in three years) caused quite a ripple in the startup world. Notably, as it imploded, the company described itself as a “trailblazer,” saying that not all such parties make it to “the mountain top,” claiming that while it failed, the startup managed to “forever” change the world of online commerce. While the debacle paled in comparison to what would come later in the year when it came to overly confident leaders (ahem, see below), it was perhaps one of the earliest signs that all was not as rosy as it appeared in fintech land.

FTX

Image Credits: Tom Williams/CQ-Roll Call, Inc / Getty Images

We debated on whether to include cryptocurrency exchange FTX as it technically has not shut down. But as one staffer pointed out, “We certainly lost it as the company it was.” The once-third-largest crypto exchange FTX on November 11 filed for bankruptcy in the U.S. and announced that CEO and founder Sam Bankman-Fried had resigned from his role. That news came days after a week-long collapse of the FTX empire as the company attempted to keep itself afloat, seeking acquisitions and fresh capital from market players. By December 12, Bankman-Fried had been arrested in the Bahamas. The next day, the U.S. Securities and Exchange Commission (SEC) had officially charged Bankman-Fried with defrauding investors.

The demise of the once-high-flying startup, which had raised nearly $2 billion in funding and once appeared to be flush with cash, no doubt marked a very low point for the crypto space. For now, Enron turnaround veteran John J. Ray III is serving as FTX’s new CEO, reportedly making $1,300 an hour.

Other crypto companies that also filed for bankruptcy this year but also technically did not shut down include Celsius and BlockFi.

Haus

Image Credits: Haus

Haus, a direct-to-consumer aperitif business backed by the likes of Casey Neistat, Homebrew Ventures and Coatue, shuttered earlier this year. What was surprising was that Haus announced this shift after it crossed the $10 million in revenue threshold and announced that it would be hitting national distribution with Winebow — two markers of growth.

Instead, the company’s eventual demise was triggered by an investor kerfuffle. Haus CEO and co-founder Helena Hambrecht said that Constellation committed to leading the startup’s $10 million Series A, and even offered to advance the startup money as runway began to dwindle. Then, last minute, Constellation backed out of the deal without any specific reasoning other than “timing,” she says.

The co-founder said “there’s no villain” in the shutdown story, yet Constellation’s dropout shows another example of how difficult it is to be a venture-backed, direct-to-consumer company.

Insteon

Proving that home automation can be a tough nut to crack, Insteon abruptly shut down in mid-April 2022, turning off its cloud servers without giving customers any warning. Launched by startup SmartLabs in 2005, Insteon at one point had an agreement with Microsoft to sell its kits at Microsoft Store locations and was one of the two launch partners for Apple’s HomeKit platform, with the HomeKit-enabled Insteon Hub Pro.

Insteon for the first few days didn’t respond to questions about the shutdown and its CEO, Rob Lilleness, deleted his LinkedIn account. Subsequently, however, the company updated its website with a statement that blamed the sudden liquidation on pandemic and supply chain problems. Apparently — if the unattributed statement is to be believed, at least — the goal was to find a parent for Insteon. But while a sale was expected in March, the plans ultimately fell through.

Insteon’s proprietary protocol likely didn’t do it any favors. More widely compatible technologies like Zigbee, Z-Wave and Matter are licensable and widely adopted, giving Insteon little in the way of leverage.

Kite

Image Credits: Kite

Kite, a startup developing an AI-powered coding assistant, shut down in November despite securing tens of millions of dollars in venture capital backing. Kite struggled to pay the bills, founder Adam Smith revealed in a postmortem blog post, running into engineering headwinds that made finding a product-market fit essentially impossible.

“We failed to deliver our vision of AI-assisted programming because we were 10+ years too early to market, i.e., the tech is not ready yet,” Smith said. “Our product did not monetize, and it took too long to figure that out.”

Kite’s failure doesn’t necessarily bode well for the other companies pursuing — and attempting to commercialize — generative AI for coding. Smith estimated that it could cost over $100 million to build a “production-quality” tool capable of synthesizing code reliably. That said, Kite’s rivals, including GitHub, Tabnine and DeepCode, believe it’s premature to become bearish on the market.

Kitty Hawk

Sebastian Thrun at TechCrunch Disrupt SF 2017. Image Credits: TechCrunch

Kitty Hawk had understandably high hopes when it launched in 2010. Founded by and piloted by self-driving car pioneer Sebastian Thrun, the eVTOL maker had some prominent backers, including, most notably, Google co-founder, Larry Page. In September, the startup announced its closure courtesy of a curt tweet, noting, “We have made the decision to wind down Kittyhawk. We’re still working on the details of what’s next.”

What comes next still isn’t entirely clear. Plenty of folks remain bullish on the eVTOL category, but Kitty Hawk couldn’t stick the landing. After flying 111 of its crafts a total of 25,000 flights, the firm shuttered that specific program, ultimately resulting in 70 layoffs. Further progress was made, “by 2022, however, the mission was less clear,” as Kirsten notes in her news report. A commercial air taxi was apparently still in the works by the time the company began winding down operations in September.

Modsy

Image Credits: Bryce Durbin / TechCrunch

In late June,Modsy, on online interior design services startup,abruptly ceased offering design services, laid off its designers and left customers with unfinished renovations and project orders in process. By July, Modsy had shut down entirely — a surprising turn of events for a startup that raised $72.7 million from investors including Comcast Ventures and NBCUniversal. So what went wrong?

Modsy took a major bottom-line hit on the logistics side during the pandemic as global supply chains ground to a halt. Amanda Kwan-Rosenbush, the former senior director of finance and accounting at Modsy, described shipping as a “significant cost” and said that Modsy’s furniture and décor partners often struggled with long delays.

But the e-design platform space is a tough nut to crack. Rivals like Laurel & Wolf and Homepolish shuttered in 2019, while Décor Aid, a smaller company, closed up shop in 2021.

Modsy made a series of aggressive cuts two years prior to its shutdown, slashing designer pay and reducing both salaried employees and its network of designers. Business of Home’s reporting revealed that the startup — in addition to piloting its own furniture line — at one point experimented with outsourcing design work to the Philippines and Bulgaria as a way to reduce operating expenses. But the pivots weren’t enough in the end to prevent Modsy’s demise.

NopeaRide

NopeaRide, Kenya’s first fully electric vehicle service, shut down in November after scaling to 70 vehicles and building a charging network all across Nairobi. It closed after parent company EkoRent Oy was unable to raise additional funding.

The closure came after the startup raised an undisclosed amount of funding since its 2018 launch. It was seeking to build more solar charging hubs in Nairobi and expand the radius in which it operated within.

Onward Mobility

Image Credits: Bryce Durbin / TechCrunch

Some startup failures are unexpected from the outside. Others you can see coming from a mile away. In spite of once sharing a blog post titled, “Contrary to popular belief, we are not dead,” Onward Mobility wasn’t fooling anyone. The Austin-based firm entered the mobile scene with an already risky proposition: bringing the BlackBerry back once again. The titular firm behind the original line struggled for years and TCL’s revival didn’t last particularly long.

Onward promised things would be different this time. It announced its intentions to the world, fell completely silent for some time and less than two years later, admitted that rumors of its death were no longer greatly exaggerated. That news arrived approximately one month after the company was publicly insisting otherwise. It’s frankly extremely hard to launch a brand new company even when there isn’t a global pandemic. And it seems like a fairly safe bet that, 15 years after the first iPhone turned the market upside down, there just isn’t enough of an appetite in the U.S. to serve as the foundation of a brand new phone maker.

Reali

Real estate fintech startup Reali began its shutdown in August in a surprise move, considering it had just raised $100 million one year prior. After a boom in home buying, the real estate tech sector found itself struggling as inflation and mortgage interest rates climbed, leading to a major slowdown in the housing market.

Even as it was winding down, Reali described itself as “one of the pioneering companies to offer the ‘buy before you sell’ and ‘cash offer’ programs to homeowners.” It seems that even being a pioneer doesn’t guarantee success and the news left us — and our readers — wondering how companies can burn through so much cash, so fast.

ShopX

Image Credits: Halfpoint Images / Getty Images

India-based ShopX filed for bankruptcy in August after failing to generate enough cash flow and running into challenges raising capital. The startup, which provided software to connect brands, retailers and in-person shoppers, had raised over $66 million in funding from Fung Group, NB Ventures and others, and was last valued at about $175 million.

ShopX competed mainly with business-to-business vendors such as 1K Kirana Bazaar and SuperK but ventured into the business-to-consumer space in 2021, offering incentives — including cash back and cash-saving offers — to customers while they browsed their neighborhood kirana shops. (In India, “kirana” are small independently owned shops that make up a major part of India’s physical retail economy.) ShopX also rewarded purchases on select bike and car-related services, salon visits, grocery, medicines and more.

Udayy

Image Credits: Boris Zhitkov (opens in a new window) / Getty Images

Edtech has had a rough year. That rings especially true for Udayy, which shut downafter raising millions from investors, reported the Economic Times. The Indian edtech sold live learning courses to kids, a use case that isn’t as bright as it used to be. As Natasha has said in the past, we now know that the startups that most enjoyed a pandemic-era boom are now the same startups facing difficult questions about how to navigate a not-so-looming downturn. The same venture capital rounds that allowed companies to expand their idea of what a total addressable market could look like, are the same tranches that may have forced an overspending and overhiring spree that now requires a correction.

Remembering the startups we lost in 2022 by Brian Heater originally published on TechCrunch

Vercel makes it easier to collaborate on preview deployments

Vercel, the popular front-end development platform, now allows you to comment directly on preview deployments, using a highly visual commenting system that is similar to Figma’s. The company had been testing this feature since September and it’s now generally available to all users.

As Vercel co-founder and CEO Guillermo Rauch told me, the idea here was to allow easier collaboration on top of the company’s existing development infrastructure, all while broadening the number of people who could get involved in the process.

Image Credits: Vercel

“The big vision behind this was that we wanted to do for front-end development, what Figma and Google Docs did for enabling this multiplayer collaboration on top of existing ways that people were working, whether it was documents or image files,” he said.

One of Vercel’s earliest innovations was that whenever there was a pull request, Vercel would give you a URL that would show the current state of the front-end application. This new commenting system builds on top of that by adding a new UI element to these previews that allows you to add comments and see existing ones. Since these previews come directly from Vercel, the company can integrate this natively, without the need for a browser extension (though there is also a Chrome extension to enable an additional feature that allows you to add screenshots to your comments).

Image Credits: Vercel

For the most part, the new commenting system is pretty straightforward. There are emoji reactions and all of the other features you’d expect from a commenting system in 2022, including, for example, an inbox that aggregates the most recent comments. As Vercel CTO Malte Ubl told me, the company at one point even enabled the ability to doodle on a page. That didn’t make it into the final product, but it may make it into a future release.

Ubl noted that the tool’s beta testers tended to use it in bursts. “You don’t necessarily have something that you want to review in this way every single day. That’s not the way developers work,” he said. “There might be changes that are not graphical at all. Then you don’t need this product and your pull request review is perfectly fine. But then, companies ramp up to launch and we see them in our statistics going from a comment thread here or there to 100 a day.”

The new commenting feature is rolling out to all Vercel users today.

Vercel makes it easier to collaborate on preview deployments by Frederic Lardinois originally published on TechCrunch

This VC is bullish about American dynamism – ‘the real one’

For many Americans, COVID-19 served as a wake-up call: The U.S. was nowhere nearly as prepared to overcome challenges as it should be, and it was overly dependent on China.

Whether the current geopolitical order calls for a new wave of isolationism will depend on your worldview, but the desire for the U.S. to be more self-reliant, solve its most glaring issues and stay ahead of its rivals can be found across the political spectrum.

Will the answer come from the government? Venture capital firm Andreessen Horowitz doesn’t think so. Its co-founder, Marc Andreessen, already said as much in his “Time to Build” call to action, but general partner Katherine Boyle doubled down in January with an essay on American dynamism.

American dynamism, Boyle wrote, is “the recognition that seemingly insurmountable problems in our society — from national security and public safety to housing and education — demand solutions that aren’t simply incremental changes that perpetuate the status quo. These problems demand solutions from builders — and it’s never been more vital that startups tackle these serious American problems.”

As a thesis, American dynamism found resonance in the U.S. tech community. But voices also emerged to question how well this goal was represented in a16z’s portfolio, which is heavier in crypto and SaaS than in govtech, for instance.

Boyle already anticipated some of that criticism, attributing it in part to a false belief “that software only touches the digital world and that we’re wasting valuable talent ignoring the physical.”

But, she retorted, “these critics are missing this current movement to the physical sector and the interplay between hardware and software.” In terms of sectors, Boyle added, “dynamism is neither govtech nor ESG.”

This is where things become interesting: a16z came up with a concept that clearly resonates with other investors, but that they may also define differently. For instance, solo GP Nichole Wischoff recently told TechCrunch that she is hoping to fund “American dynamism … the real one.”

This VC is bullish about American dynamism – ‘the real one’ by Anna Heim originally published on TechCrunch

OpenAI releases Point-E, an AI that generates 3D models

The next breakthrough to take the AI world by storm might be 3D model generators. This week, OpenAI open-sourced Point-E, a machine learning system that creates a 3D object given a text prompt. According to a paper published alongside the code base, Point-E can produce 3D models in 1 to 2 minutes on a single Nvidia V100 GPU.

Point-E doesn’t create 3D objects in the traditional sense. Rather, it generates point clouds, or discrete sets of data points in space that represent a 3D shape — hency the cheeky abbreviation. (The “E” in Point-E is short for “efficiency,” because it’s ostensibly faster than previous 3D object generation approaches.) Point clouds are easier to synthesize from a computational standpoint, but they don’t capture an object’s fine-grained shape or texture — a key limitation of Point-E currently.

To get around this limitation, the Point-E team trained an additional AI system to convert Point-E’s point clouds to meshes. (Meshes — the collections of vertices, edges and faces that define an object — are commonly used in 3D modeling and design.) But they note in the paper that the model can sometimes miss certain parts of objects, resulting in blocky or distorted shapes.

Image Credits: OpenAI

Outside of the mesh-generating model, which stands alone, Point-E consists of two models: a text-to-image model and an image-to-3D model. The text-to-image model, similar to generative art systems like DALL-E 2 and Stable Diffusion, was trained on labeled images to understand the associations between words and visual concepts. The image-to-3D model, on the other hand, was fed a set of images paired with 3D objects so that it learned to effectively translate between the two.

When given a text prompt — for example, “a 3D printable gear, a single gear 3 inches in diameter and half inch thick” — Point-E’s text-to-image model generates an synthetic rendered object that’s fed to the image-to-3D model, which then generates a point cloud.

After training the models on a data set of “several million” 3D objects and associated metadata, Point-E could produce colored point clouds that frequently matched text prompts, the OpenAI researchers say. It’s not perfect — Point-E’s image-to-3D model sometimes fails to understand the image from the text-to-image model, resulting in a shape that doesn’t match the text prompt. Still, it’s orders of magnitude faster than the previous state-of-the-art — at least according to the OpenAI team.

Converting the Point-E point clouds into meshes.

“While our method performs worse on this evaluation than state-of-the-art techniques, it produces samples in a small fraction of the time,” they wrote in the paper. “This could make it more practical for certain applications, or could allow for the discovery of higher-quality 3D object.”

What are the applications, exactly? Well, the OpenAI researchers point out that Point-E’s point clouds could be used to fabricate real-world objects, for example through 3D printing. With the additional mesh-converting model, the system could — once it’s a little more polished — also find its way into game and animation development workflows.

OpenAI might be the latest company to jump into the 3D object generator fray, but — as alluded to earlier — it certainly isn’t the first. Earlier this year, Google released DreamFusion, an expanded version of Dream Fields, a generative 3D system that the company unveiled back in 2021. Unlike Dream Fields, DreamFusion requires no prior training, meaning that it can generate 3D representations of objects without 3D data.

While all eyes are on 2D art generators at the present, model-synthesizing AI could be the next big industry disruptor. 3D models are widely used in film and TV, interior design, architecture and various science fields. Architectural firms use them to demo proposed buildings and landscapes, for example, while engineers leverage models as designs of new devices, vehicles and structures.

Point-E failure cases.

3D models usually take a while to craft, though — anywhere between several hours to several days. AI like Point-E could change that if the kinks are someday worked out, and make OpenAI a respectable profit doing so.

The question is what sort of intellectual property disputes might arise in time. There’s a large market for 3D models, with several online marketplaces including CGStudio and CreativeMarket allowing artists to sell content they’ve created. If Point-E catches on and its models make their way onto the marketplaces, model artists might protest, pointing to evidence that modern generative AI borrow heavily from its training data — existing 3D models, in Point-E’s case. Like DALL-E 2, Point-E doesn’t credit or cite any of the artists that might’ve influenced its generations.

But OpenAI’s leaving that issue for another day. Neither the Point-E paper nor GitHub page make any mention of copyright.

OpenAI releases Point-E, an AI that generates 3D models by Kyle Wiggers originally published on TechCrunch

Apple fixes bug that let malicious apps skirt macOS’ security protections

Microsoft says a vulnerability it discovered in a core macOS security feature, Gatekeeper, could have allowed attackers to compromise vulnerable Macs with malware.

The flaw, tracked as CVE-2022-42821, was first uncovered by Microsoft principal security researcher Jonathan Bar Or, and dubbed the “Achilles” vulnerability. Bar Or said the bug could allow malware to skirt Gatekeeper’s protections on macOS.

First introduced in 2012, Gatekeeper is a security feature designed to allow only trusted software to run on macOS. The feature automatically verifies that all apps downloaded from the internet are from identified developers who have been “notarized” by Apple, and whose apps are known to be free of malicious content.

Microsoft’s Bar Or explained in a blog post that macOS adds a “quarantine” attribute to apps and files that have been downloaded from a web browser and instructs Gatekeeper to check the file before it can be opened. But the Achilles vulnerability exploits a file permissions model called Access Control Lists (ACLs) to add extremely restrictive permissions to a downloaded file, which prevents web browsers from properly setting the quarantine attribute.

In exploiting the bug, a user could be tricked into downloading and opening a malicious file on macOS without triggering Gatekeeper’s security protections.

Microsoft reported the Achilles flaw in July, but Apple didn’t acknowledge the vulnerability was fixed until last week.

Bar Or said that Lockdown Mode, an opt-in Apple feature introduced earlier this year to help high-risk users block some of the more sophisticated cyberattacks, would not defend against the Achilles vulnerability, since Lockdown Mode is aimed at stopping silent and remotely triggered “zero-click” attacks that require no user interaction. “End-users should apply the fix regardless of their Lockdown Mode status,” said Bar Or.

Achilles is just one of many Gatekeeper bypasses that have been uncovered in recent years. In April 2021, Apple fixed a zero-day vulnerability in macOS that enabled the threat actors behind the notorious Shlayer malware to bypass Apple’s Gatekeeper and notarization security checks.

Apple fixes bug that let malicious apps skirt macOS’ security protections by Carly Page originally published on TechCrunch

Amazon and EU settle two antitrust cases, including one focused on merchant data abuse

The European Commission (EC) has announced that it has reached an agreement with Amazon over a duo of antitrust probes, one that will enshrine commitments made by Amazon in European Union (EU) antitrust legislation.

The initial probe kicked off back in 2018, when regulators launched enquiries into how Amazon was leveraging non-public data from third-party marketplace sellers on its platform to benefit its own competing business as a retailer. The crux of the concerns centered on how Amazon was able to gain an unfair advantage through big data insights as the marketplace owner, such as optimizing its own pricing or deciding what new products to launch and when.

The probe escalated into a formal investigation the following year, before the EC issued Amazon with a direct Statement of Objections in 2020. Europe’s competition chief Margrethe Vestager said at the time that Amazon was likely abusing its market position in its biggest European markets in France and Germany, and was “illegally distorting” competition through its use of merchant data.

At the same time, the EC announced a second tangential investigation into how Amazon favored its own business in terms of rules it set merchants for being featured in its much-coveted “buy box” and Prime loyalty program. The Commission added that Amazon seemingly favored its own products, as well as sellers that use Amazon’s logistics and delivery services.

Commitments

In the intervening months, Amazon submitted proposals to appease regulators in an attempt to end the probe early, including commitments to: stop using non-public data from its marketplace sellers; treat all sellers equally, regardless of whether they pay for Amazon’s logistics services; allow Prime sellers to choose any carrier for their deliveries. However, the bloc was urged by NGOs, trade unions, and digital rights groups to reject what they deemed to be a “weak” offer by the ecommerce giant, arguing that the Commission should pursue the probe through to its natural conclusion, which may eventually have involved a huge fine.

Fast-forward to today, and the Commission has said that Amazon has made some amendments to its initial offer, which includes improving the layout of a second competing “buy box” that Amazon had earlier proposed, and several other changes that it says will increase transparency and data protection for third-party merchants on the platform.

“Today’s decision sets new rules for how Amazon operates its business in Europe,” noted Margrethe Vestager, the European Commission’s executive vice-president for competition policy, in a statement. “Amazon can no longer abuse its dual role and will have to change several business practices. Competing independent retailers and carriers as well as consumers will benefit from these changes opening up new opportunities and choice.”

These commitments, according to the EC, will be legally binding and cover Amazon’s activities across the whole European Economic Area (EEA), though Italy is excluded from the “buy box” and Prime commitments due to a separate case brought by Italy against Amazon back in 2021.

The Prime and “buy box” commitments will remain enforceable for seven years, while all the remaining commitments will apparently lapse after five years. If Amazon is found to have breached any part of these commitments in that period, it could face a fine of 10% of its global revenue.

Today’s announcement comes just a day after the EC issued a preliminary finding that Facebook’s parent company Meta abused its dominant market position in the classifieds ads space, in contravention ofArticle 102 of the Treaty on the Functioning of the European Union (TFEU), the same treaty that Amazon allegedly contravened.

However, while Amazon will have to change the way it operates in Europe, the company said that it doesn’t agree with a number of the EC’s assertions.

“We are pleased that we have addressed the European Commission’s concerns and resolved these matters,” an Amazon spokesperson told TechCrunch in a statement. “While we continue to disagree with several of the preliminary conclusions the European Commission made, we have engaged constructively to ensure that we can continue to serve customers across Europe and support the 225,000 European small and medium sized businesses selling through our stores.”

Amazon and EU settle two antitrust cases, including one focused on merchant data abuse by Paul Sawers originally published on TechCrunch

Pin It on Pinterest