South Africa’s Flow gets funding to automate social media advertising for real estate agencies

The process used by millions of agents and thousands of property portals globally to reach buyers and sellers on digital channels is highly fragmented. And it’s evident that proptech, unlike other industries, has lagged in utilizing social media to make sales.

South African startup Flow wants to change how real estate agencies, developers and agents interact with their end customers. With its APIs, Flow connects to the websites of estate agencies and property developers and automates advertising for them on social media channels like Instagram and Facebook. The proptech marketing platform is announcing that it has raised $4.5 million in pre-Series A funding.

The proptech intends to use the funding to include other social media platforms such as TikTok and LinkedIn and other advertising channels like out-of-home billboards. The investment will see co-founders and co-CEOs Gil Sperling and Daniel Levy drive the business’s B2B growth strategy and integrate Flow’s social media–driven real estate marketing platform into existing international property portals and CRM platforms.

Sperling and Levy founded Flow in 2017 as an app that rewards tenants for early rent payments. However, before Flow, both founders previously built an adtech and performance marketing company, Popimedia, which was the largest buyer of Facebook media inventory in Africa for some of the world’s biggest brands. While they sold the business to global communications group Publicis in 2016, it was some of the knowledge gained while running Popimedia that they drew on to pivot Flow into its current business model four years later.

“With our first adtech business, we never dealt with real estate or property as we could never really service them in this country [South Africa]. And the biggest problem was that as much as real estate is the biggest asset class in the world and very valuable vertical, it is the least innovated around because it’s just highly fragmented,” Sperling told TechCrunch on a call.

“When buying and selling homes, if you take South Africa, for example, 40,000 agents are marketing 300,000 listings at any time. Every agent is essentially a little small business because they’re commissioners, and there’s no way that they can afford to each have a marketing, data science department, and design department like big businesses can, and that is one reason why we couldn’t conduct ads or performance marketing for many of them.”

With Flow, the founders want real estate agencies and property developers they couldn’t reach with their former startup to connect with customers on digital channels. The proptech startup automates the marketing for real estate agents for developers and works hand-in-hand with real estate websites such as Property24 and Private Property to pull listings and automatically create ads on Facebook, Instagram and other digital channels.

According to Flow, its proptech marketing platform improves revenue for agents and experiences for property buyers and sellers. On the other hand, Levy points out that the startup makes money when these agents use its SaaS platform and via a percentage cut from their marketing spend. He added that revenue has been growing 20% month-on-month within the past year.

“Our route to market, for the most bit, has been going door-to-door from franchisor to franchisee to different offices within that group. And over the last couple of months, we’ve identified the enterprise channel, as we call it, which is more associated with strapping on our technology to portals,” stated the co-CEO. “So our next phase of traction and growth will come from those relationships, which are significant in our world. And that’s why we’ve just gone through this capital raise to experiment with that essentially.”

Flow currently has over 300+ clients using its platform — a client being a real estate agency or developer where each office has about 15 to 20 smaller agents. So more broadly, Flow is used by nearly 6,000 agents across South Africa, Namibia, Botswana, Mauritius and Australia. It is in talks with partners, mainly property portals and CRM platforms, to expand into Europe (France, Germany, Belgium and the U.K.) where it’ll face stiffer competition — which the co-founders hope Flow will edge out with its technology and attention to design — but a more extensive market base.

Futuregrowth Asset Management led Flow’s pre-Series A round with $2 million. Endeavor Harvest Fund and serial entrepreneur Steven Heilbron participated, while existing investors Kalon Venture Partners, Vunani Fintech Fund and Buffet Investments also doubled down.

“We’ve keenly followed Flow’s progress in South Africa and Australia and integration into the B2B side of the global property industry as the next natural step in the company’s evolution,” says Futuregrowth Asset Management head of Private Equity and Venture Capital, Amrish Narrandes, on the investment. “We share Daniel and Gil’s vision to bring the property industry into the 21st century and know they have the expertise and experience to make it happen — and we’re pleased to be able to be part of a South African company taking bold steps that will bring much-needed change to an essential global industry.”

South Africa’s Flow gets funding to automate social media advertising for real estate agencies by Tage Kene-Okafor originally published on TechCrunch

That Microsoft deal isn’t exclusive, video is coming, and more from OpenAI CEO Sam Altman

OpenAI co-founder and CEO Sam Altman sat down for a wide-ranging interview with this editor late last week, answering questions about some of his most ambitious personal investments, as well as about the future of OpenAI.

There was much to discuss. The now eight-year-old outfit has dominated the national conversation in the two months since it released ChatGPT, a chatbot that answers questions like a person. OpenAI’s products haven’t just astonished users; the company is reportedly in talks to oversee the sale of existing shares to new investors at a $29 billion valuation despite its comparatively nominal revenue.

Altman declined to talk about the company’s current business dealings, firing a bit of a warning shot when asked a related question during our sit-down.

He did reveal a bit about the company’s plans going forward, however. For one thing, in addition to ChatGPT and the outfit’s popular digital art generator, DALL-E, Altman confirmed that a video model is also coming, though he said that he “wouldn’t want to make a competent prediction about when,” adding that “it could be pretty soon; it’s a legitimate research project. It could take a while.”

Altman made clear that OpenAI’s evolving partnership with Microsoft — which first invested in OpenA in 2019 and earlier today confirmed it plans to incorporate AI tools like ChatGPT into all of its products — is not an exclusive pact.

Further, Altman confirmed that OpenAI can build its own software products and services, in addition to licensing its technology to other companies. That’s notable to industry watchers who’ve wondered whether OpenAI might one day compete directly with Google via its own search engine. (Asked about this scenario, Altman said: “Whenever someone talks about a technology being the end of some other giant company, it’s usually wrong. People forget they get to make a counter move here, and they’re pretty smart, pretty competent.”)

As for when OpenAI plans to release the fourth version of the GPT, the sophisticated language model off which ChatGPT is based, Altman would only say that the hotly anticipated product will “come out at some point when we are confident that we can [release] it safely and responsibly.” He also tried to temper expectations regarding GPT-4, saying that “we don’t have an actual AGI,” meaning artificial general intelligence, or a technology with its own emergent intelligence, versus OpenAI’S current deep learning models that solve problems and identify patterns through trial and error.

“I think [AGI] is sort of what is expected of us” and GPT-4 is “going to disappoint” people with that expectation, he said.

In the meantime, asked about when Altman expects to see artificial general intelligence, he posited that it’s closer than one might imagine but also that the shift to “AGI” will not be as abrupt as some expect. “The closer we get [to AGI], the harder time I have answering because I think that it’s going to be much blurrier and much more of a gradual transition than people think,” he said.

Naturally, before we wrapped things up, we spent time talking about safety, including whether society has enough guardrails in place for the technology that OpenAI has already released into the world. Plenty of critics believe we do not, including worried educators who are increasingly blocking access to ChatGPT owing to fears that students will use it to cheat. (Google, very notably, has reportedly been reluctant to release its own AI chatbot, LaMDA over concerns about its “reputational risk.)

Altman said here that OpenAI does have “an internal process where we kind of try to break things and study impacts. We use external auditors. We have external red teamers. We work with other labs and have safety organizations look at stuff.”

At the same time, he said, the tech is coming — from OpenAI and elsewhere — and people need to start figuring out how to live with it, he suggested. “There are societal changes that ChatGPT is going to cause or is causing. A big one going on now is about its impact on education and academic integrity, all of that.” Still, he argued, “starting these [product releases] now [makes sense], where the stakes are still relatively low, rather than just put out what the whole industry will have in a few years with no time for society to update.”

In fact, educators — and perhaps parents, too — should understand there’s no putting the genie back in the bottle. While Altman said that OpenAI and other AI outfits “will experiment” with watermarking technologies and other verification techniques to help assess whether students are trying to pass off AI-generated copy as their own, he also hinted that focusing too much on this particular scenario is futile.

“There may be ways we can help teachers be a bit more likely to detect output of a GPT-like system, but honestly, a determined person is going to get around them, and I don’t think it’ll be something society can or should rely on long term.”

It won’t be the first time that people have successfully adjusted to major shifts, he added. Observing that calculators “changed what we test for in math classes” and Google rendered the need to memorize facts far less important, Altman said that deep learning models represent “a more extreme version” of both developments. But he argued the “benefits are more extreme as well. We hear from teachers who are understandably very nervous about the impact of this on homework. We also hear a lot from teachers who are like, ‘Wow, this is an unbelievable personal tutor for each kid.’”

For the full conversation about OpenAI and Altman’s evolving views on the commodification of AI, regulations, and why AI is going in “exactly the opposite direction” that many imagined it would five to seven years ago, it’s worth checking out the clip below.

You’ll also hear Altman address best- and worst-case scenarios when it comes to the promise and perils of AI. The short version? “The good case is just so unbelievably good that you sound like a really crazy person to start talking about it,” he’d said. “And the bad case — and I think this is important to say — is, like, lights out for all of us.”

That Microsoft deal isn’t exclusive, video is coming, and more from OpenAI CEO Sam Altman by Connie Loizos originally published on TechCrunch

Gogoro, Belrise JV to spend $2.5B on battery swapping network in Indian state

Gogoro, the Taiwanese paladin for two-wheeler battery swapping, is working with the Indian state of Maharashtra to establish state-wide battery charging and swapping infrastructure. The initiative will be driven by a “50-50” partnership between Gogoro and Belrise Industries, an automotive systems manufacturer based in India, that will form an infrastructure company to own the batteries and swapping stations in Maharashtra, according to Gogoro’s CEO and co-founder Horace Luke.

Additional infrastructure-related investors will participate down the line in the joint venture, which aims to invest up to $2.5 billion over eight years in Maharashtra, according to a non-binding MOU signed by the state, Gogoro and Belrise at the World Economic Forum in Davos.

“Like most infrastructure deployments, the partnership company will raise funding for the smart battery swapping buildout,” Luke told TechCrunch in response to a question on whether Gogoro would use its free cash flow to fund the initiative.

Additionally, Gogoro will also form an India-based company that will run the battery swapping network, a spokesperson told TechCrunch.

Per the MOU, Gogoro and Belrise will start installing swapping stations in the top 10 cities of Maharashtra, starting with Mumbai in the next few months, and expand beyond that in the future.

Maharashtra is one of India’s largest commercial and industrial centers. As a result, the state has some of the best installed electricity generation capacity in India, with around a quarter of its power mix coming from wind and solar capacity. Gogoro says Maharashtra has an energy surplus, making it a good launch point for a connected network.

The joint venture is one of the latest of such initiatives to be announced in Maharashtra. In September, Mumbai-based battery swapping startup VoltUp partnered with Adani Electricity and Hero Electric to set up 500 electric mobility stations in Mumbai over the next two years. Last June, Sun Mobility, a provider of energy infrastructure and services for EVs, announced the launch of its own battery-swapping network for EVs in Mumbai in collaboration with Amazon India. Sun plans to deploy more than 2,000 battery swapping stations across the region by 2025.

Gogoro didn’t disclose how many swapping stations it aims to build over the next eight years, but Luke said based on the population of Maharashtra (about 120 million), the network there will be about four times the size of Gogoro’s network in Taiwan. To date, Gogoro has 12,200 battery swapping stations in Taiwan, with more than 1.1 million smart batteries in circulation throughout its network. We can therefore infer that Gogoro plans to build somewhere in the ballpark of 50,000 stations in the region.

Taiwan’s total build-out has cost Gogoro $640 million over the last seven years, a spokesperson told TechCrunch, so $2.5 billion will bring about unprecedented scale.

“Battery swapping is a new category creation and we will build out the network infrastructure to induce demand,” said Luke.

A big part of Gogoro’s formula for inducing demand is to provide vehicles that are built with its own battery swapping technology. In Taiwan, Gogoro’s battery swapping network powered 90% of all electric scooters in 2023, including five out of the top six electric vehicle makers. Gogoro won’t be bringing its own brand of scooters to India, but the company is working with Hero MotoCorp, a popular two-wheeled vehicle maker in India, to launch electric two-wheelers based on Gogoro’s technology under the Hero brand.

That partnership was announced two years ago, but no vehicles have shipped yet. Which makes sense for Gogoro’s business model. The company wants people to use its swapping stations, so installing the publicly available swapping infrastructure before widespread rollout of electric two-wheelers is a signal that Gogoro is marking its territory.

Gogoro, Belrise JV to spend $2.5B on battery swapping network in Indian state by Rebecca Bellan originally published on TechCrunch

P2P lending platform PeopleFund raises $20M Series C extension led by Bain Capital

PeopleFund, a South Korean marketplace that connects borrowers and investors to enable lending, has recently added $20 million to its $63.4 million Series C.

Existing backer Bain Capital led the extension, with participation from previous investors such as Access Ventures, CLSA Capital Partners Lending Ark Asia, D3 Jubilee Partners, 500 Global, Kakao Investment, TBT Partners and IBX Partners.

The additional funding brings PeopleFund’s total raised to around $100 million in equity. Apart from the capital, PeopleFund also secured $240 million in debt financing in 2022 from Goldman Sachs, CLSA Lending Ark Asia and Bain Capital. The company did not disclose its valuation when asked.

In 2021, PeopleFund raised $63.4 million (75.9 billion won) in equity for Series C, also led by Bain Capital, to further develop its credit-scoring system.

PeopleFund plans to use its new capital to continue to advance its AI-powered risk management and credit scoring system for its users, which includes borrowers and lenders. On top of that, the startup aims to launch a B2B service this year to provide AI-enabled customized credit scoring system services to financial institutions.

Another reason for its runway extension is to meet one of the requirements for a P2P lending license, according to industry sources. In South Korea, P2P lending marketplaces must pass yearly requirements to get a license from Financial Services Commission (FSC) to run their business. To operate its business in 2023, PeopleFund, which reports that it is making a profit loss, must own a minimum capital ranging from $400,000 to $2.4 million, depending on its loan balance. (The loan balance is the remaining amount of loans made by PeopleFund that the borrowers have not yet repaid.) PeopleFund’s loan balance was $264.3 million (326.8 billion won) as of December 2022, the company said. That means the outfit’s requirement capital is around $1.5 million to $2.4 million, according to the industry sources and local media.

Joey Kim, founder of PeopleFund, said in a statement that “2022 will be marked as a year of turbulence for fintech, with the global public market adjustment alongside changes in the macro environment. Meanwhile, the Korean consumer lending market has undergone a dramatic transition into the mobile sphere, with big players like KakaoPay and Toss leading the change. This transition, coupled with the instability of the credit market, is opening up opportunities for tech-based digital lenders and its technologies to highlight our competence compared to traditional financial institutions.”

Image Credits: PeopleFund

The outfit says its total amount of loans deployed to borrowers to date was estimated at $1.3 billion in December, up from $936 billion in October 2021. The startup says it has seen more than 56.7% growth in the number of borrowers and 9.6% in the number of lenders compared to the previous year. The number of its borrowers and lenders was 20,688 and 2,943,883, respectively, as of December last year.

The Seoul-based P2P lending startup, founded in 2015, successfully closed its extension. Still, the impact of the extremely tough market condition was inevitable, leading to several tech industry layoffs in the last few months. PeopleFund confirmed that it had cut about 10% of its staff in the fourth quarter of 2022 to “operate the business efficiently and effectively” amid the possibility of a worsening economy. PeopleFund had nearly 150 people as of December 2021.

Want more fintech news in your inbox? Sign up here.

P2P lending platform PeopleFund raises $20M Series C extension led by Bain Capital by Kate Park originally published on TechCrunch

Daily Crunch: Apple powers up 14- and 16-inch MacBook Pro models with M2 Pro and Max chips

To get a roundup of TechCrunch’s biggest and most important stories delivered to your inbox every day at 3 p.m. PST, subscribe here.

Good news — we are back with another round of events for TechCrunch, and we have early-bird tickets available to our Early Stage event in Boston in April. Woohoo!

Today, we also particularly enjoyed Connie’s peek into the future as Sam Altman sees it. — Christine and Haje

The TechCrunch Top 3

An Apple a day…: Get ready to take a big bite out of Apple. Kyle and Brian peel off the cover to unveil the consumer tech giant’s M2 Pro and M2 Max chips that pack some power. The chips will be available in the 14- and 16-inch MacBook Pros, and if you are looking around for a new computer, the M2 Mac Mini arrives January 24, starting at $599.
Cloudy with a chance of embedded analytics: Nothing but blue skies so far for Cumul.io, a low-code embedded analytics platform for SaaS companies. The company raised $10.8 million to continue developing its business intelligence platform that quickly “connects just about any data source, drag-and-drop specific features to customize their dashboards, and then copy-paste a snippet of code into their application to serve thousands of end-users,” Paul writes.
The headline says it all: We couldn’t help but steal from the TC+ section because Tim’s headline is just so good: “Nest co-founder Matt Rogers’ new startup is trash.” We won’t ruin it any more for you.

Startups and VC

ChatGPT, the AI that can write poems, emails, spreadsheet formulas and more, has attracted a lot of negative publicity lately, Kyle writes. That’s perhaps why AI21 Labs, an Israeli startup developing text-generating AI systems along the lines of ChatGPT, tried a different tack with its newly released assistive writing tool, Wordtune Spices. A part of AI21’s expanding suite of generative AI, Wordtune Spices doesn’t compose emails and essays like ChatGPT. Instead, it suggests options that change the voice and style of already written sentences, also offering up statistics from web-based sources to “strengthen arguments.”

Apropos robots that write…On the heels of raising at a $1 billion valuation last week, DeepL is taking the wraps off a new language product, the first extension for a startup that made its name from its popular AI-based translation tools, Ingrid, er, writes. Write is a new tool that fixes your writing — catching grammar and punctuation mistakes, offering suggestions for clarity and more creative phrasing and (soon) giving you the option to change your tone.

There was a lot of fun startup news on the site over the past few days, so it was hard to choose just five to load up in our little recommendation engine, but here’s what we came up with:

QR isn’t dead…: It’s the tech that won’t die, as Beaconstac lands a $25 million investment for its QR code management platform, Kyle reports.
The DM says nah: Amanda reports that D&D publisher says “We rolled a 1” as it addresses backlash over controversial license.
That price is un-Hinge-d: Dating app Hinge tests a pricier $60 per month subscription, similar to Tinder Platinum, Sarah writes.
Want to buy that again?: CloseFactor raises $15.2 million to automate repetitive sales processes, Kyle reports.
Time for 100 push-ups, stat: Darrell takes a closer look at Vitruvian’s Trainer+, writing that it is an all-in-one home gym that actually lives up to its promises.

7 space tech predictions for 2023

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

At the time of this writing, Wikipedia notes that there have been eight successful spaceflight launches so far this year.

New spaceports are entering operation, cell phone users will soon have connectivity from space, and the Artemis program backed by NASA is one of several ventures that will bring robots (and eventually human crews) to the moon.

“Despite the economic uncertainty, we believe new records will be established in spacetech as giant commercial projects get funded,” says Mark Boggett, CEO and co-founder of Seraphim Space Manager LLP.

Three more from the TC+ team:

A feature does not a company make: Build a company, not a feature, by Haje.
Mo IP, mo money: Cost-effective IP strategies can lead to massive exit valuations, by Kyle Graves.
Here’s a flying crystal ball: Mark Boggett has 7 space tech predictions for you for the upcoming year.

TechCrunch+ is our membership program that helps founders and startup teams get ahead of the pack. You can sign up here. Use code “DC” for a 15% discount on an annual subscription!

Big Tech Inc.

Extra, extra, read all about how the Royal Mail confirmed a cyberattack that disrupted postal service in the United Kingdom. Carly writes that the confirmation comes a week after the Royal Mail said it was hit by an unspecified “cyber incident” that caused it to not be able to dispatch items overseas. CEO Simon Thompson said he didn’t believe customer data was compromised, but notified authorities in case that changes. Some reports say the LockBit ransomware group is behind this, and Carly is working on confirming that.

And we have five more for you:

“Baby Yoda” back again: Get ready for more cuteness from a galaxy far, far away. Disney+ released ‘The Mandalorian’ Season 3 trailer, Lauren writes. May the Force be with you.
We see you: Frederic writes that Wyze is going back to its OG way with the Wyze Cam OG and OG Telephoto security cameras.
Do you hear what I hear?: Ivan and Jagmeet paired up to report on a new pilot Google is doing of a “soundbox” in India for merchants to get audio-based payment alerts.
If you like fast cars and you cannot lie: Get ready to burn rubber. Chevy unveiled its fastest Corvette yet, the electrified 2024 Corvette E-Ray. Matt has more.
One company’s trash is another company’s treasure: If you’re thinking of getting a giant neon Twitter bird light for your home or office, Amanda knows where you can procure one.

Daily Crunch: Apple powers up 14- and 16-inch MacBook Pro models with M2 Pro and Max chips by Christine Hall originally published on TechCrunch

Tesla engineer testifies that 2016 video promoting self-driving was faked

Tesla faked a 2016 video promoting its self-driving technology, according to testimony by a senior engineer reviewed by Reuters.

The video, which shows a Tesla Model X driving on urban, suburban and highway streets; stopping itself at a red light; and accelerating at a green light is still on Tesla’s website and carries the tagline: “The person in the driver’s seat is only there for legal reasons. He is not doing anything. The car is driving itself.”

CEO Elon Musk used the video as evidence that Tesla “drives itself” by relying on its many built-in sensors and self-driving software. Yet according to Ashok Elluswamy, director of Autopilot software at Tesla, the video was staged using 3D mapping on a predetermined route, a feature that is not available to consumers.

In his July deposition, which was taken as evidence in a lawsuit against Tesla for a fatal 2018 crash involving former Apple engineer Walter Huang, Elluswamy said Musk wanted the Autopilot team to record “a demonstration of the system’s capabilities.”

Elluswamy’s statement also confirms and provides more details on what anonymous former employees told the New York Times in 2021. While there appeared to be no legal ramifications for Tesla following the NYT’s investigation, an on-the-record testimony from a current employee could cause trouble for the automaker, which is already beleaguered by lawsuits and investigations surrounding its Autopilot and Full Self-Driving (FSD) systems. (To be clear, neither system is actually self-driving. They are advanced driving-assistance systems that automate certain driving tasks, but as Tesla has made clear on its website, drivers should stay alert and keep their hands on the steering wheel when the systems are engaged.)

When electric truck maker Nikola was accused of and eventually admitted to faking a video of its fuel cell–powered Nikola One semitruck prototype — Nikola had actually placed the truck on a small hill, allowing gravity, not the motor, to do its thing — state and federal investigations were launched into both Nikola and its chairman and founder, Trevor Milton. Milton was found guilty on charges of securities fraud in October.

Tesla’s fake video was created using 3D mapping on a predetermined route from a house in Menlo Park, California, to Tesla’s office in Palo Alto, according to Elluswamy. Drivers had to intervene to take control during test runs, and the scenes that were left on the cutting room floor included the test car crashing into a fence in Tesla’s parking lot when trying to park itself without a driver.

“The intent of the video was not to accurately portray what was available for customers in 2016. It was to portray what was possible to build into the system,” Elluswamy said, according to a transcript of his testimony seen by Reuters.

Musk promoted the video at the time, tweeting Tesla vehicles require “no human input at all” to drive through urban streets to highways and eventually to find a parking spot.

Neither Musk nor Tesla, which has disbanded its press office, responded in time to TechCrunch’s request for comment.

The revelation comes at a time when Tesla is facing litigation for multiple fatal crashes involving its Autopilot system, as well as a criminal investigation from the U.S. Department of Justice for claims Tesla made about Autopilot. Just this week, a Tesla that had FSD engaged suddenly accelerated and crashed into a BC Ferries ramp in Canada, totaling the vehicle.

Regarding the 2018 crash that killed Huang, the National Transportation Safety Board concluded in 2020 Tesla’s “ineffective monitoring of driver engagement” had contributed to the crash, which the board said was likely caused by Huang’s distraction and the limitations of the system.

While Tesla does tell its drivers to pay attention to the road, there are ways drivers can fool the system to make the car believe they were paying attention, said Elluswamy. Many drivers even go so far as to buy Tesla counterweights on websites like Alibaba, which can be placed on steering wheels to mimic the weight of human hands that are otherwise engaged while the car is in movement.

Even amid regulator scrutiny and reports of crashes, Tesla recently extended access of its FSD software to customers across North America.

Tesla engineer testifies that 2016 video promoting self-driving was faked by Rebecca Bellan originally published on TechCrunch

Discord acquires Gas, a compliments-based social media app for teens

The messaging platform Discord announced its acquisition of Gas, an app that’s popular among teens for its positive spin on social media.

On Gas, users sign up with their school, add friends, and answer polls about their classmates. But the questions in the polls are intended to boost users’ confidence, rather than damage it. Teens might be asked to choose which of four friends is the best DJ, or has the best smile. Then, the person who was chosen will get an anonymous message with their compliment, sent from a vague “boy in 10th grade” or “girl in 11th grade.”

Gas was founded by Nikita Bier, who previously sold a similar app called tbh to Facebook in 2017 — tbh has since been shut down.

According to data from Sensor Tower, Gas reached 7.4 million installs and almost $7 million in consumer spending since its launch in summer 2022. Users can subscribe to a paid feature called “God Mode,” which gives users hints about who their secret complimenters are.

Gas got gassed up on the Today Show this morning https://t.co/qEJfMp7M2Z

— Nikita Bier (@nikitabier) January 10, 2023

“At this time, Gas will continue as its own standalone app and the Gas team will be joining Discord to help our efforts to continue to grow across new and core audiences,” Discord wrote in an announcement. As of October, Bier said Gas had four team members.

Despite Gas’s fast popularity, the app has had a rocky road to its exit. The app was the subject of a widespread sex trafficking rumor, which was completely false, yet still impacted the app’s downloads. Bier told the Washington Post that he and his team received hundreds of graphic death threats as a result of this hoax. Other viral social apps IRL and WalkSafe have also been hit with unfounded trafficking accusations.

Gas is one of several anonymous apps — some based around compliments, some not — that have gone viral recently. But TechCrunch found that apps like NGL and Sendit were using bots to simulate engagement. Like Gas, these apps offer users the ability to pay to see who asked questions. Understandably, some customers felt scammed when it turned out that these questions didn’t actually come from their friends. Meanwhile, 9count, the company that made Spark and Summer, is working on a product similar to Gas called nocapp.

Even though Discord is currently going to keep Gas operating as a standalone product, Discord recently announced that it would integrate a selection of apps into its servers. So, it’s possible that we could see these positive community polls on the messaging platform in the future.

“We’re always working to create an inclusive world where no one feels like an outsider and we’re excited to welcome Gas to the Discord community as our next step to fulfilling that vision,” Discord wrote in its announcement.

The terms of Discord’s acquisition of Gas have not been disclosed.

Discord acquires Gas, a compliments-based social media app for teens by Amanda Silberling originally published on TechCrunch

Oversight Board presses Meta to revise ‘convoluted and poorly defined’ nudity policy

Meta’s Oversight Board, which independently evaluates difficult content moderation decisions, has overturned the company’s takedown of two posts that depicted a non-binary and transgender person’s bare chests. The case represents a failure of a convoluted and impractical nudity policy, the Board said, and recommended that Meta take a serious look at revising it.

The decision concerned two people who, as part of a fundraising campaign for one of the couple who was hoping to undergo top surgery (generally speaking the reduction of breast tissue). They posted two images to Instagram, in 2021 and 2022, both with bare chests but nipples covered, and included a link to their fundraising site.

These posts were repeatedly flagged (by AI and users) and Meta ultimately removed them, as violations of the “Sexual Solicitation Community Standard,” basically because they combined nudity with asking for money. Although the policy is plainly intended to prevent solicitation by sex workers (another issue entirely), it was repurposed here to remove perfectly innocuous content.

When the couple appealed the decision and brought it to the Oversight Board, Meta reversed it as an “error.” But the Board took it up anyway because “removing these posts is not in line with Meta’s Community Standards, values or human rights responsibilities. These cases also highlight fundamental issues with Meta’s policies.”

They wanted to take the opportunity to point out how impractical the policy is as it exists, and to recommend to Meta that it take a serious look at whether its approach here actually reflects its stated values and priorities.

The restrictions and exceptions to the rules on female nipples are extensive and confusing, particularly as they apply to transgender and non-binary people. Exceptions to the policy range from protests, to scenes of childbirth, and medical and health contexts, including top surgery and breast cancer awareness. These exceptions are often convoluted and poorly defined. In some contexts, for example, moderators must assess the extent and nature of visible scarring to determine whether certain exceptions apply. The lack of clarity inherent in this policy creates uncertainty for users and reviewers, and makes it unworkable in practice.

Essentially: even if this policy did represent a humane and appropriate approach to moderating nudity, it’s not scalable. For one reason or another, Meta should modify it. The summary of the Board’s decision is here and includes a link to a more complete discussion of the issues.

The obvious threat Meta’s platforms face, however, should they relax their nudity rules, is porn. Founder Mark Zuckerberg has said in the past that making his platforms appropriate for everyone necessitates taking a clear stance on sexualized nudity. You’re welcome to post sexy stuff and link to your OnlyFans, but no hardcore porn in Reels, please.

But the Oversight Board says this “public morals” stance is likewise in need of revision (this excerpt from the full report lightly edited for clarity):

Meta’s rationale of protecting “community sensitivity” merits further examination. This rationale has the potential to align with the legitimate aim of “public morals.” That said, the Board notes that the aim of protecting “public morals” has sometimes been improperly invoked by governmental speech regulators to violate human rights, particularly those of members of minority and vulnerable groups.

…Moreover, the Board is concerned about the known and recurring disproportionate burden on expression that have been experienced by women, transgender, and non-binary people due to Meta’s policies…

The Board received public comments from many users that expressed concern about the presumptive sexualization of women’s, trans and non-binary bodies, when no comparable assumption of sexualization of images is applied to cisgender men.

The Board has taken the bull by the horns here. There’s no sense dancing around it: the policy of recognizing some bodies as inherently sexually suggestive, but not others, is simply untenable in the context of Meta’s purportedly progressive stance on such matters. Meta wants to have its cake and eat it too: give lip service to people like the trans and NB people like those who brought this to its attention, but also respect the more restrictive morals of conservative groups and pearl-clutchers worldwide.

The Board Members who support a sex and gender-neutral adult nudity policy recognize that under international human rights standards as applied to states, distinctions on the grounds of protected characteristics may be made based on reasonable and objective criteria and when they serve a legitimate purpose. They do not believe that the distinctions within Meta’s nudity policy meet that standard. They further note that, as a business, Meta has made human rights commitments that are inconsistent with an approach that restricts online expression based on the company’s perception of sex and gender.

Citing several reports and internationally-negotiated definitions and trends, the Board’s decision suggests that a new policy be forged that abandons the current structure of categorizing and removing images, substituting something more reflective of modern definitions of gender and sexuality. This may, of course, they warn, leave the door open to things like nonconsensual sexual imagery being posted (much of this is automatically flagged and taken down, something that might change under a new system), or an influx of adult content. The latter, however, can be handled by other means that total prohibition.

When reached for comment, Meta noted that it had already reversed the removal and that it welcomes the Board’s decision. It added: “We know more can be done to support the LGBTQ+ community, and that means working with experts and LGBTQ+ advocacy organizations on a range of issues and product improvements.” I’ve asked for specific examples of organizations, issues, or improvements and will update this post if I hear back.

Oversight Board presses Meta to revise ‘convoluted and poorly defined’ nudity policy by Devin Coldewey originally published on TechCrunch

Putting numbers on the global venture slowdown

“Tell me something I don’t know,” was my first reaction when reports on venture capital in 2022 hit my inbox this month. It is pretty clear by now that there was a slowdown, so what’s the point of harping on about it as we enter a new year?

The point, as often with data, is that we can now confirm what was merely intuition until Q4 actually closed: 2022 saw fewer exits and venture deals than 2021.

The Exchange explores startups, markets and money.

Read it every morning on TechCrunch+ or get The Exchange newsletter every Saturday.

If we had to retain only one fact, it would be the decline in capital invested in startups last year. According to CB Insights, whose State of Venture report is one of our sources today, total venture funding in 2022 amounted to $415.1 billion, 35% less than in 2021.

According to the PitchBook-NVCA Venture Monitor report, deal count was more stable, with 2022’s estimated deal count approaching 2021’s figure. But looking at quarterly data reveals that Q4 had the lowest deal count of 2022, which doesn’t necessarily bode well for 2023.

Putting numbers on the global venture slowdown by Anna Heim originally published on TechCrunch

2022 global smartphone shipments were the lowest in nearly a decade

One of these days I’ll have some positive news to share about the global smartphone market. Today is not that day. The industry capped off another dismal year with a 17% year over year drop for Q4. That number puts the full year’s shipping figures 11% below 2021, per new numbers from Canalys, which refer to it as “an extremely challenging year for all vendors.”

It’s been one thing after another from the industry. Slowing figures pre-dated 2020, while the pandemic and its various knock-on effects have continued tossing up roadblocks. For 2022, the same macroeconomic headwinds that have impacted practically every facet of life took their own toll on the industry. Notably, the figures for the quarter and the year were at their lowest in nearly a decade. The firm tells TechCrunch, “we have to go back to 2013 to find lower numbers — and back then the market situation was very different as the technology was a lot more emerging.”

Image Credits: Canalys

Apple returned to the top spot for Q4, at a quarter of the total market. Samsung held onto No. 2, but still captured the top spot for the entirety of 2022.

“The channel is highly cautious with taking on new inventory, contributing to low shipments in Q4,” analyst Runar Bjørhovde said in a new release. “Backed by strong promotional incentives from vendors and channels, the holiday sales season helped reduce inventory levels. While low-to-mid-range demand fell fast in previous quarters, high-end demand began to show weakness in Q4. The market’s performance in Q4 2022 stands in stark contrast to Q4 2021, which saw surging demand and easing supply issues.”

It’s a long way of saying the industry saw a bit of a rebound when supply chain constraints began to ease up, but additional external forces reversed those positive trends — and then some. The firm doesn’t expect much in the way of rebound for the remainder of 2023, either, predicting that growth will be “flat to marginal,” as economic uncertainty and inflation remains.

Unemployment, interest rate hikes and other issues are expected to have an adverse effect on “mid-to-high-end-dominated markets,” including North America and parts of Europe. That should, to some extent, be counteracted by a bump from China, as the world’s largest smartphone market continues the reopening process.

2022 global smartphone shipments were the lowest in nearly a decade by Brian Heater originally published on TechCrunch

Pin It on Pinterest