TechCrunch+

In 2018, appliance conglomerate Bosch created a startup, Security and Safety Things (or “SAST” for short), whose stated mission was to develop a platform to help developers create software for AI-equipped cameras. SAST was to host a moderated, vetted “app store” for internet-connected cameras that would allow developers to build software on an open standard — software mainly focused on security and “business intelligence” use cases.

SAST successfully launched the app store in 2020, later rebranding it (and itself) to Azena and opening a headquarters in Pittsburgh’s Strip District. But after tens of millions of euros in investment from Bosch, SAST — now Azena — apparently never quite achieved the success that its parent company hoped it would.

TechCrunch has learned that Azena is shutting down its external operations and pivoting to internal projects at Bosch. In a statement, a Bosch spokesperson said that partners and customers have been informed and that Azena will “fully honor” its existing contractual obligations.

“Moving forward, Azena will focus on Bosch internal business and stop external business development. This includes a transition to maintenance and support only for [Azena’s software],” the spokesperson said via email. “All platform components of Azena stay operational for now … We are actively working on a transition plan.”

Azena’s marketplace was relatively robust by IP camera market standards, with around 100 apps at its peak. Like popular app stores for smartphones, it let developers sell their apps to customers and provide demos for pilot projects. The app store would handle backing up and restoring settings and ensuring configurations remained consistent across cameras.

Prior to its partial shutdown, Azena had also been developing an operating system for cameras that enabled supported models to run multiple AI-enabled apps simultaneously. Built on top of Android, manufacturers — including Qisda/Topview, AndroVideo, Vivotek and Bosch itself — sold cameras running the firmware, which powered apps for heat mapping and queue analysis in retail stores, automated payment processing, license plate recognition and more.

As of September 2021, Azena had over 120 employees spread throughout its offices in Munich, its facility in Pittsburgh and its R&D hub in Eindhoven, The Netherlands. The startup counted NHL hockey team the Pittsburgh Penguins among its customers, who used the Azena platform to monitor crowding at stadium entrances, recognize license plates and identify overcrowding near fan merchandise retail locations.

Azena generated controversy earlier this year when it came to light that the startup was only carrying out basic auditing of the software hosted in its app store. According to the company’s terms of use, responsibility for the ethics and legality of the apps rested squarely on the shoulders of developers and users. Some apps claimed to accurately detect weapons and analyze human behavior, applications that many ethicists say are beyond the capabilities of even the most sophisticated AI systems.

In a public response at the time, Azena noted that it required developers working on its platform to commit to abiding by ethical business standards laid out by the United Nations. But the startup admitted that it didn’t have the ability to check how Azena-powered cameras were used and didn’t verify whether apps sold on its store were legal or in compliance with developer agreements.

An investigation by the Intercept also found evidence that Azena was years behind on patching security exploits that could allow hackers access to cameras running its operating system. Azena disputed the suggestion but acknowledged that Azena’s firmware permitted users to sideload apps outside of the app store onto supported cameras.

Bosch shuts down its app store for AI-powered, internet-connected cameras by Kyle Wiggers originally published on TechCrunch

Nnamdi Okike and Aaron Holiday trust data over the kind of pattern matching that most VCs swear by. It’s not surprising, given their backgrounds. Before launching their venture firm, 645 Ventures, in 2014, Okike was a principal with the data-driven investment giant Insight Partners. Meanwhile, Holiday, who came directly from DFJ Gotham Ventures, was previously a software engineer at Goldman Sachs.

Of course, data is hard to come by when a startup is just getting off the ground. But last week, in an exchange with TechCrunch, Okike and Holiday said that their proprietary software and “resource-intensive model to early-stage investing” is working so well that 645 just secured $347 million in capital commitments from a range of traditional venture investors (foundations, family offices, endowments) across two new funds. One is a $195 million early-stage fund; the other is a $153 million fund to back its breakout winners as they mature.

These are notable amounts in a market where LPs are feeling a lot less flush than they did a year ago. The funds are a far cry, too, from the duo’s $7.6 million proof-of-concept fund. (They went on to raise $40 million from LPs in 2018 and $160.6 million in 2020, so they now manage around $555 million altogether.)

Some liquidity from FiscalNote, a maker of policy management software that began trading publicly in August after merging with a blank-check company, presumably helped. The team says seven other portfolio companies have meanwhile sold and that many others are far more valuable than when 645 Ventures backed them, including two outfits whose later rounds were led by Insight. These include the $60 million Series C of the national medical practice Eden Health and the $50 million Series B round of cybersecurity company Shift5.

Other milestones to crow about (for now) include the $1.4 billion valuation of the cloud security firm Panther Labs and Overtime, a sports league startup that recently raised a $100 million growth round.

Of course, there have been misses. Okike says that one big regret is the NFT marketplace OpenSea, which he and Holiday had an opportunity to invest in at the seed stage and “unfortunately passed.” At the time, he says, they “didn’t appreciate how fast the NFT market was growing.”

The good news, suggests Holiday, is that there are plenty of other great New York–based startups to fund. “As a member of the board at Cornell Tech, we’re seeing lots of young entrepreneurial energy flock to the city, and several tech founders have relocated their headquarters from other tech hubs to NYC.”

The team also sees it as positive that more “storied Silicon Valley VCs” who “once viewed San Francisco as the center of the universe” are opening offices in New York, including, earlier this year, both Index Ventures and Sequoia Capital.

Are they worried at all about the competition? Not really, says Okike. “We have tailored our offering to provide a compelling value proposition as a lead investor.”

One of the firm’s newest bets is Efficient Capital Labs, which lends capital from U.S. firms to software companies in India; 645 Ventures led its $3.5 million seed round.

Okike says the firm also closed a not-yet-disclosed Series A investment in a real estate software startup and that it has been, and will remain, active in real estate software, with past bets that include RentSpree, a startup focused on online rental applications and tenant screening; Aryeo, a real estate content platform that helps its users manage, syndicate, and automate listing content; and Rifiniti, a business intelligence startup that was acquired in 2019 by FM:Systems, a maker of facilities management software, for undisclosed terms.

A data-driven duo just raised roughly $350M to fund seed-stage startups with metrics by Connie Loizos originally published on TechCrunch

Meta employees were told that they should not discuss sensitive issues like abortion, gun control, pending legislation and vaccine efficacy at work. Fortune reported on these changes, citing a leaked internal memo from Lori Goler, head of people at Meta. TechCrunch confirmed the report with a Meta spokesperson.

“As Mark mentioned recently, we need to make a number of cultural shifts to help us deliver against our priorities,” Goler wrote in a company memo, per Fortune’s report. “We’re doing this to ensure that internal discussions remain respectful, productive, and allow us to focus. This comes with the trade-off that we’ll no longer allow for every type of expression at work, but we think this is the right thing to do for the long-term health of our internal community.”

Meta took a similar stance in June, when a draft Supreme Court opinion that would overturn Roe v. Wade was leaked. According to a document that the New York Times obtained at the time, Meta said that “discussing abortion openly at work has a heightened risk of creating a hostile work environment.”

“We deeply value expression, open discussion and a company culture built on respect and inclusivity,” said Kadia Koroma, a Meta spokesperson, in an email to TechCrunch. “We’ve updated our employee expectations to provide direction around what is appropriate for our people in the workplace, so that we can reduce distractions while maintaining an environment that is respectful and inclusive and where people can do their best work.”

Meta employees who are required to discuss these topics to do their jobs are exempt from the policies. These guidelines don’t extend outside of the workplace.

As a company, Meta is in a period of financial tumult as its metaverse investments fail to pay off. At the beginning of the year, Meta stock traded at around $330 a share; now, it’s dropped by about 50% to $115 per share. Over the summer, CEO Mark Zuckerberg told employees in an all-hands call that he would ramp up expectations and set more aggressive goals. “Realistically, there are probably a bunch of people at the company who shouldn’t be here,” he told his team. Then, last month, Meta cut 11,000 jobs, amounting to 13% of its workforce.

These new mandates to avoid discussion about sensitive issues fall in line with Zuckerberg’s desire to increase intensity at work. The changes are positioned as a way to keep employees focused by “minimizing disruption,” Goler’s note says, per Fortune. Goler also addressed the way that Meta takes positions on public policy.

“We are often asked to sign on to advocacy letters on topics that are important, but not directly connected to our work. This can distract us from focusing on issues that are not central to our mission,” Goler wrote. “So going forward, as a company we will only make public statements on issues that are core to our business, meaning they are required in order to provide our service.”

Coinbase took a similar approach in 2020, as CEO Brian Armstrong posted a culture memo stating that discussions of political issues and social causes were not allowed — if employees didn’t like it, they could take severance and leave.

The policy was controversial. For many tech workers, especially those from underrepresented backgrounds, current events have a palpable impact on their day-to-day lives — and that includes work. Even Jack Dorsey, former Twitter CEO and Bitcoin evangelist, spoke out against Coinbase’s anti-activism policy, writing that crypto is “direct activism against an unverifiable and exclusionary financial system which negatively affects so much of our society.” Dorsey said that Armstrong’s stance “leaves people behind.”

At Meta, a company operating social media platforms that billions of people use every day, it’s hard to imagine that these banned topics won’t inevitably come up.

Meta won’t let staff discuss topics like abortion, gun control and vaccines at work by Amanda Silberling originally published on TechCrunch

Fintech was hot in 2021, but looking back on it … maybe too hot?

The sector exploded last year, seeing record investment — $132 billion globally, according to CB Insights — with many startups reaching lofty valuations, including Stripe at $95 billion, Klarna at $45 billion and Plaid at $13 billion. While these companies have very real customer bases and products, it is not hard to imagine that at least some of these valuations were propped up by hype.

The dominoes have already started to fall here. In July, Swedish buy now, pay later startup Klarna raised $800 million at a new $6.7 billion valuation. That marks a stunning 85% decrease in valuation over the course of roughly a year. Ouch.

But it was at least refreshing that Klarna’s CEO, Sebastian Siemiatkowski, was one of the few who didn’t shy away from the realities of startup valuations in this market. He took to Twitter after Klarna announced its new lower valuation to acknowledge the current market conditions and state that the lower valuation didn’t mean it was actually doing all that much worse, citing profitability and growth into new markets. Klarna has continued to launch into new geographies since.

So, how are the other fintech hotshots from last year doing? Well, not so hot.

Fintech unicorn valuations have fallen hard in 2022 by Rebecca Szkutak originally published on TechCrunch

Vietnamese electric vehicle maker VinFast has filed for an initial public offering in the United States, the company said Tuesday. Shares will be listed on the Nasdaq under the ticker “VFS.”

VinFast, which was founded in 2017 and began operations in 2019, will convert to a Singapore public limited company for the IPO. The number of shares to be offered and the price range of the offering haven’t been disclosed.

The EV startup has been pursuing the U.S. market, most recently with a showcase of four SUVs presented at the LA Auto Show. Over the summer, VinFast received $1.2 billion in incentives to build a factory in North Carolina, where the automaker hopes to begin building cars by July 2024. VinFast has even promised a $7,500 discount to potential American buyers that would hold out to buy an EV eligible for U.S. EV tax incentives.

No date was given for the IPO, which was originally slated for Q4 of this year. It’s more likely we’ll see the company go public sometime next year, given current market uncertainty.

Unlike many EV companies that have chosen to go public through a special purpose acquisition merger, VinFast has already begun producing and shipping vehicles. The automaker shipped its first batch of 999 vehicles to the U.S. late last month.

Vietnamese luxury EV-maker VinFast files to go public on Nasdaq by Rebecca Bellan originally published on TechCrunch

We recently uncovered that Lensa AI can be tricked into creating NSFW images. When TechCrunch made the Prisma team aware of its findings, the company’s CEO replied with its findings.

Prisma Lab’s CEO and co-founder Andrey Usoltsev told us that the behavior we observed in our article can only happen if the AI is intentionally provoked into creating NSFW content, and points out that this represents a breach against its terms of use.

“Our Terms of Use (Clause 6) and Stability AI Terms of Service (Prompt Guidelines) explicitly prohibit the use of the tool to engage in any harmful or harassing behavior. The way [TechCrunch’s] experiment was structured points out that such creations can’t be produced accidentally. The images are the result of intentional misconduct on the app,” said Usoltsev in an email interview with TechCrunch. “Generation and wide usage of such content may incur legal actions, as both the US and the UK regard an act of sharing of explicit content and imagery generated without consent as a crime. We provide guidelines that clearly stipulate our image requirements and the use of any explicit depictions is strictly prohibited. We expect the app’s users to follow the guidelines to receive the best possible results.”

Usoltsev also shared some additional context for why Lensa ended up generating NSFW images, explaining that this is a result of the underlying technology, Stability AI, is doing what it is told, but only in a sandbox environment.

“Stable Diffusion neural network is running behind the avatar generation process,” says Usoltsev. “Stability AI, the creators of the model, trained it on a sizable set of unfiltered data from across the internet. Neither us, nor Stability AI could consciously apply any representation biases; To be more precise, the man-made unfiltered data sourced online introduced the model to the existing biases of humankind. The creators acknowledge the possibility of societal biases. So do we.”

The Stability AI model includes adaptations of the Stable Diffusion model software to make it harder for users to generate nude and pornographic imagery since the end of November, 2022, and Prisma AI’s founder assures me that these adaptations can be outmanouvered by savvy users.

“We specify that the product is not intended for minors and warn users about the potential content. We also abstain from using such images in our promotional materials,” Usoltsev told us. “To enhance the work of Lensa, we are in the process of building the NSFW filter. It will effectively blur any images detected as such. It will remain at the user’s sole discretion if they wish to open or save such imagery.”

The Prisma Labs team points out that there are two issues here; by uploading explicit images, the users train a particular and individual copy of the model, that the company claims is deleted once the generation is complete, and that these images cannot be used to train the model further. In other words: If you upload porn to make more porn, that’s kind of on you.

“There’s no doubt that a wider conversation around AI use and regulations needs to take place in the near future and we’re keen to be a part of it. We also provide all necessary guidance and appropriate warnings to enable the best experience of the Magic Avatars feature,” says Usoltsev. “But if an individual is determined to engage in harmful behavior, any tool would have the potential to become a weapon.”

The company didn’t share whether it has plans in place to avoid the creation of so-called ‘deepfake’ nude imagery.

In the meantime, I guess I get to enjoy consensually-generated photos of myself looking better than I ever have in any photo, and encourage others to obtain consent before they create porn of others.

Prisma Labs, maker of Lensa AI, says it is working to prevent accidental generation of nudes by Haje Jan Kamps originally published on TechCrunch

Reading the spreadsheet detailing the investment portfolio of Alameda Research, the investment arm of fallen crypto exchange FTX, you wonder how they had time to do anything other than invest given the sheer number of deals recorded. Perhaps that was part of the problem.

FTX and its sister company (or parent company, depending on how you look at it) Alameda had their hands in a bunch of different startups. The depth of its roster wasn’t very transparent until now.

A spreadsheet first shared by the Financial Times showed Alameda’s private equity portfolio, with some FTX positions included. The document includes just shy of 500 investments across 10 holding companies for a total of $5.276 billion. (Like the Financial Times, TechCrunch has yet to confirm all data shared in the spreadsheet, meaning that when we discuss aggregates, we’re speaking directionally. We reached out to FTX and its founder, Sam Bankman-Fried, for comment but haven’t heard back.)

This spreadsheet, dated from early November, raises a number of concerns surrounding the extent to which FTX and Alameda — and their affiliated companies — invested in the crypto industry.

“I scratched my head at the FTX investments/acquisitions (i.e. Dave Inc/Storybook) last year and thought maybe SBF (as a genius) saw the market differently, and maybe I was losing my touch,” Vance Spencer, co-founder of Framework Ventures, tweeted on Tuesday. “Looking at it all together in 2022: nope, they were idiots, they lit all the money on fire.”

FTX and Alameda’s massive investments will take a long time to unwind from crypto industry by Jacquelyn Melinek originally published on TechCrunch

It’s been a pretty rough week for Salesforce co-founder and CEO Marc Benioff and the folks at his company: Three talented executives – co-CEO Bret Taylor, Tableau CEO Mark Nelson, and Slack CEO and co-founder Stewart Butterfield – announced their resignations in quick succession.

It’s fair to ask what exactly is going on at Salesforce to lose three accomplished people so quickly, but it’s also important to parse each exit to determine whether they are part of a political battle or just some odd confluence of unconnected events.

The news seems to have spooked investors, with the company’s stock down nearly 17% over the last five days. But what do these departures mean to Salesforce and to the companies it spent so much money to acquire over the last several years? Further, how does it impact the executive depth that Benioff has worked so hard to build up? Finally, does he look for another co-CEO to help him run the company, or does he continue running it alone for the foreseeable future?

And another one gone, another one gone

Let’s start with Nelson. He’s the least well-known of the three. Salesforce bought his company, Tableau, in 2019 for almost $16 billion. At the time, the company was run by Adam Selipsky, who left last year to become CEO at AWS when Andy Jassy was promoted to Amazon CEO after Jeff Bezos stepped back from that role

For every action, there is an equal and opposite reaction in the C-suite, apparently.

As Butterfield exits stage left, it’s fair to wonder what’s happening at Salesforce by Ron Miller originally published on TechCrunch

Apple has talked a big game about its future plans to break into the automotive market. But a report from Bloomberg says that Apple has had to scale back its plans. Last year, Apple said it would debut a fully self-driving car in 2025, but now, the vehicle is delayed until 2026 and will not be autonomous.

Apple sought to build the first completely autonomous vehicle, without the need for a steering wheel or pedals. But according to the report, engineers working on the project (known internally as Titan) no longer think that vision is possible with current technology.

Apple has experienced a number of executive shakeups on the Titan team, which could have influenced this delay. Last year, Ford Motor snagged Doug Field, the engineering executive who was in charge of this special projects team at Apple. A few months before that, several other key players on the Titan project departed the company.

Even companies with a long history in the automotive space have struggled to make the dream of autonomous driving a reality. Of course, Tesla has also strived toward this feat with its misleadingly named “full self-driving” feature, which launched in beta in October 2020 and is now the subject of a Department of Justice criminal investigation. Launched last year, the inquiry was initiated following over a dozen accidents involving the active use of Tesla’s Autopilot system, some resulting in fatalities.

Meanwhile, TechCrunch reported in October that Argo AI, the autonomous vehicle unicorn with backing from Ford and Volkswagen, would be shutting down. Ford later said in its earnings report that it would focus more on advanced driver assistance systems than self-driving technology.

Apple’s first car, delayed until 2026, won’t be self-driving by Amanda Silberling originally published on TechCrunch

Google announced today that it’s making it easier for users to drill down on a search and explore related topics. Search currently has a few filters to help you refine and separate your search results between videos, news, images or shopping results. Now, the search giant is going to start showing users a scrollable list of related topics alongside its current filters at the top of the search results page.

For example, if you’re searching for dinner ideas, you might see filter topics like “healthy” or “easy” pop up in the new scrollable list. If you tap on one of the filters, it will add it to your search query. You can add or remove topics, which are designated by a + symbol, to quickly zoom in or backtrack on a search.

Topics are dynamic and will change as you tap in order to give you more options and help you explore new areas, Google says. For instance, if you tapped on a “healthy” filter, you may see “vegetarian” or “quick” appear next. Google says the new change will roll out for English users in the U.S. on iOS, Android and the web in the coming days.

“When you conduct a search, our systems automatically display relevant topics for you based on what we understand about how people search and from analyzing content across the web,” Google said in a blog post. “Both topics and filters are shown in the order that our systems automatically determine is most helpful for your specific query. If you don’t see a particular filter you want, you can find more using the “All filters” option, which is available at the end of the row.”

Today’s announcement comes a day after Google introduced a “Continuous Scrolling” experience on desktop so users don’t have to navigate across pages to find relevant search results, expanding a feature the company has offered on mobile for some time.

Google Search’s new topic filters make it easier to refine results or expand searches by Aisha Malik originally published on TechCrunch