Trump EPA targets two-man geoengineering startup for ‘polluting the air’
Humans have found it hard to quit fossil fuels, which is why some argue that we’ll soon need to start geoengineering — that is, modifying the atmosphere to prevent catastrophic warming of the planet.
The practice is controversial. Some argue it’s the only solution given that we’ve waited too long to reduce carbon emissions. Others say we shouldn’t be running two uncontrolled experiments on the Earth’s climate (the first being the global burning of fossil fuels).
That hasn’t stopped people from trying. And one approach championed by Make Sunsets has drawn the attention of the U.S. Environmental Protection Agency.
The startup is basically two guys from Silicon Valley who have been releasing weather balloons filled with hydrogen gas and sulfur dioxide particles. When the balloon floats somewhere past 66,000 feet in altitude, it bursts and releases the sulfur dioxide particles, which scatter and reflect sunlight, cooling the Earth a tiny bit.
The company sells “cooling credits” based on how much estimated warming each balloon release negates. Make Sunsets has raised $750,000, according to PitchBook, and the startup says its investors include Boost VC, Draper Associates, and Pioneer Fund.
Neither founder is a scientist, but the science behind sulfur dioxide and solar reflectivity is sound. Humans accidentally proved the importance of sulfur dioxide in global albedo — the average reflectivity of the Earth’s surface — when they slashed the sulfur content of marine shipping fuels in 2020; one prominent climate scientist has argued in favor of the practice.
Still, given the complexity of the global climate, it’s not clear what other effects the practice might have. It might seed rainstorms in one region while depriving other areas of rain. Several scientists have urged caution.
Plus, if sulfur dioxide particles drift closer to ground, they could aggravate people’s asthma and cause other respiratory problems. Here, the EPA takes issue with Make Sunsets’ approach to geoengineering. Sulfur dioxide is regulated as an air pollutant. EPA Administrator Lee Zeldin said this week that the agency is investigating the company.
Make Sunsets argues its actions are legal. In an FAQ on its website, the company says, “Yes, our method to cool Earth falls under the Weather Modification Act of 1976 and report yearly to NOAA of our deployments as required.”
The law is fuzzy here, though. When it was written, the Weather Modification Act was likely intended to cover the practice of cloud seeding, in which particles like silver iodide are shot into clouds to induce rain or snowfall. Most weather modification today is done by entities like ski resorts and irrigation districts in the West. It’s not clear how the law applies to climate modification.
Still, while the EPA might be justified in its investigation, it isn’t exactly consistent with Zeldin’s approach to pollution.
The agency’s efforts to boost coal are likely to generate far more sulfur dioxide pollution than Make Sunsets will release with its balloons. A Make Sunsets balloon released on November 15, 2024, released 1,715 grams of sulfur dioxide. In 2023, U.S. power plants released 650,000 tons into the atmosphere, the majority of which came from coal. That’s approximately the same amount as 343,900,000 of the startup’s balloons.
Keep reading the article on Tech Crunch
From Bogotá to the Battlefield: LatAm startups are winning big in TechCrunch Startup Battlefield
For startups around the world, the TechCrunch Startup Battlefield program offers unmatched exposure, credibility, and connections to scale their businesses. TechCrunch has long been committed to spotlighting companies solving real-world problems in scalable and sustainable ways — making it a powerful launchpad for startups both in and outside the U.S.
Several Latin American startups have thrived in this environment. Tackling critical challenges in their regions with bold, innovative solutions, these companies have turned Startup Battlefield momentum into international success. With increased visibility and valuable connections, they’ve attracted global investors, expanded into new markets, and grown their teams.
Some LatAm startups that have used Startup Battlefield as a springboard to the global stage
Salva Health (Colombia)
Salva Health is transforming breast cancer detection with its Julieta screening device, making early detection more affordable and accessible. Winning 2024 Startup Battlefield supercharged their visibility and positioned them for rapid growth.
Check out this episode of the Equity podcast to hear Salva’s founder discuss how the competition helped propel the company forward.
Ripio (Argentina)
Ripio (formerly BitPagos) is a blockchain-powered fintech company offering digital wallets, crypto trading, and decentralized credit to promote financial inclusion in Latin America. After being selected as a Startup Battlefield New York finalist in 2016, Ripio raised a $50 million Series B led by Digital Currency Group, fueling expansion into Colombia, Mexico, Uruguay, and Spain.
Jefa (Mexico)
Jefa is a digital challenger bank built for unbanked and underbanked women in Latin America, offering mobile-first services like savings accounts and debit cards. After rising to prominence as a 2020 Startup Battlefield finalist, Jefa was acquired by fintech company Tala, accelerating its mission to close the gender gap in financial access.
Olho do Dono (Brazil)
Olho do Dono uses 3D imaging to help livestock farmers in Latin America monitor cattle weight and health without the need for physical scales. After winning Startup Battlefield Latin America in 2018, the company gained regional recognition and validated its innovative approach to modernizing livestock management.
Why LatAm startups should apply
Global exposure
Being featured by TechCrunch puts startups in front of a global audience of investors, customers, and collaborators — amplifying their reach far beyond local markets.
Investor visibility
Startup Battlefield draws attention from top-tier venture capitalists, angels, and global investors who closely follow the competition. The exposure can lead to valuable conversations and funding opportunities with aligned backers.
Credibility and validation
Participation signals that a startup’s solution is globally relevant and scalable. This validation helps attract interest from media, enterprise partners, and investors alike.
Access to a powerful network
Founders join a dynamic community of fellow entrepreneurs, mentors, and industry leaders. The program fosters organic connections that often lead to partnerships, guidance, and long-term support.
Ready to be the next LatAm success story?
Apply now to Startup Battlefield and show the world what you’re building.
Keep reading the article on Tech Crunch
The Nintendo Switch 2 will still cost $450 in the US, despite tariffs
Nintendo announced on Friday that it will open Switch 2 pre-orders in the U.S. on April 24, following a delay due to the Trump administration’s hefty tariffs.
Fans feared the long-awaited console would spike in price due to Trump’s export rules on international goods, but Nintendo confirmed the Switch 2 will still cost $449.99 and come out on June 5, as originally announced.
Switch 2 accessories, however, will increase in price due to “changes in market conditions,” Nintendo said in a press release. The company added that other products could also face hikes down the road.
Most of the Switch 2 accessories seem to be getting a $5 price bump. The Nintendo Switch 2 Pro Controller, which was initially set to cost $79.99, will now cost $84.99. Joy-Cons will cost $94.99, up from $89.99, and the new Switch 2 camera attachment will cost $54.99, up from $49.99.
As far as games go, Nintendo isn’t changing the prices of the digital and physical versions of two launch-day heavy-hitters, Mario Kart World ($79.99) and Donkey Kong Bananza ($69.99).
“We apologize for the retail pre-order delay, and hope this reduces some of the uncertainty our customers may be experiencing,” Nintendo wrote in the release.
Though the changing prices of accessories may concern some customers, they won’t find reprieve from other gaming companies. Earlier this week, Sony announced that it would bump PlayStation 5 prices in Europe, Australia, and New Zealand due to shifts in the global economy. Fans expect that similar price adjustments will hit the U.S. eventually, too.
Certain electronics, like smartphones and laptops, are exempted from Trump’s tariffs, but not video game consoles.
Keep reading the article on Tech Crunch
Final day to submit your speaker application and shape the next wave of startups at TechCrunch All Stage
Today’s the day! The application to speak at TechCrunch All Stage closes tonight at 11:59 p.m. PT — this is your final chance to share real-world insights with 1,200+ startup founders and VCs attending the event.
Whether you’ve built or backed startups, battled bottlenecks, or cracked the code on growth, the stage is yours. TC All Stage lands in Boston on July 15, and we’re giving the mic to those who’ve lived the scaling grind.
Step-by-step: What to expect
Make your voice count. Your experience could help the next generation of founders grow smarter and faster. Apply now and you might land a spot in our Audience Choice round — where TechCrunch readers choose who gets the spotlight. Here’s how the process works:
Step 1: Submit your session proposal through the Call for Content form on the TC All Stage page.
Step 2: Our events programming and editorial team will carefully review every submission to identify the strongest sessions and most compelling ideas.
Step 3: Top-picked sessions will move on to the Audience Choice round, where TechCrunch readers vote for the ones they’re most excited to see at TC All Stage.
Step 4: The three sessions with the most votes will be selected to lead their own roundtables — a highly interactive discussion designed for deep engagement.
How roundtable sessions work
Roundtables are 30-minute, small-group discussions led by you and up to two speakers of your choice. These informal sessions skip the slides and video, creating space for deep, focused conversations on niche topics with highly engaged attendees.
Speaking perks
When you take the stage at TC All Stage, you don’t just get the mic — you gain full access to the entire event. Participate in roundtables with founders and investors, engage in impactful breakout sessions, and forge valuable connections with startup leaders. Perks include:
- Full event access for you and your team.
- Spotlights across TechCrunch.com, the event page, and the event app.
- Promotion on our social channels.
- Direct engagement with founders and VCs.
Don’t miss your chance to shape the conversation at TC All Stage
Help founders scale with precision, speed, and resilience. Inspire innovation, ignite growth, and influence the future of startups. Take the stage at TC All Stage and share your scaling expertise with founders and VCs eager for actionable insights. Establish yourself as a trusted leader in the startup ecosystem.
Apply before the day ends — the speaker application deadline is tonight at 11:59 p.m. PT!
Keep reading the article on Tech Crunch
Netflix is revamping search with AI to improve discovery
Netflix is building a new search experience aimed at improving the discovery experience, and it’s going to use AI to do it, the company’s CEO Greg Peters said during its first-quarter results conference call.
Peters said Netflix is working on “interactive search that’s based on generative technologies” to help people find different titles.
Answering an analyst’s question, Peters said the company’s most popular titles that dominate popular conversation draw 1% of traffic, and because of this trend, improving discovery and recommendation is important to the company.
“There is more room to improve the discovery and recommendation experience, and therefore provide more value for members, and therefore find the biggest audiences for our titles,” he said.
Earlier this month, Bloomberg reported that Netflix has been testing OpenAI-powered search with select customers in Australia and New Zealand.
The company, which also started testing a new home page design on its TV app that brought a new way to display show titles and descriptions, plans to roll it out widely later this year.
“This [Homepage redesign] is something that we hadn’t made big structural changes to in over a decade. We believe that this will significantly improve Netflix’s discovery experience. We’ve been polishing and improving that experience based on the input we got from members who used it,” Peters noted.
The company said first-quarter revenue rose 12.5% to $10.54 billion from a year earlier.
Keep reading the article on Tech Crunch
OpenAI pursued Cursor maker before entering into talks to buy Windsurf for $3B
When news broke that OpenAI was in talks to acquire AI coding company Windsurf for $3 billion, one of the first questions on the mind of anyone following the space was likely: “Why not buy Cursor creator Anysphere instead?”
After all, OpenAI Startup Fund has been an investor in Anysphere, the maker of Cursor, since the quickly-growing coding assistant’s seed round in late 2023. (Anysphere is often referred to by its product name, Cursor.) It turns out that OpenAI indeed approached Anysphere in 2024 and again earlier this year about a potential acquisition, according to a report from CNBC. The talks failed. Instead, Anysphere has been in talks to raise capital at about $10 billion valuation, Bloomberg reported last month.
OpenAI’s desire to move on to acquisition discussions with another coding assistant maker signals how important capturing a slice of the code generation market has become for the ChatGPT maker. Windsurf is generating about $40 million in annualized recurring revenue (ARR), TechCrunch reported in February. Meanwhile, Anysphere’s Cursor reportedly makes about $200 million on an ARR basis.
While OpenAI’s Codex CLI “agent,” which the company released Wednesday, can also write and edit code, its attempt to buy Windsurf suggests the company doesn’t want to wait for CLI to gain traction with customers.
Keep reading the article on Tech Crunch
Defense tech Theseus landed Y Combinator, the US Special Forces, and $4.3M from a tweet
On February 18, 2024, Ian Laffey posted on X that he and two others he’d just met built a cheap drone at a hackathon that calculated its coordinates simply by using its camera and Google Maps. He and his colleagues, Sacha Lévy and Carl Schoeller, were all engineers under the age of 25.
The tech had clear potential to combat rampant GPS jamming of drones in Ukraine. Instead of GPS, drone operators there have to use high-tech goggles to guide their drones by sight. But that leads to lots of problems, especially under poor conditions like thick fog or at night.
At the end of the hackathon, Schoeller wished his two teammates well and parted, hoping their paths might cross again.
But the tweet went viral and changed their lives. A day later, the three decided to apply to Y Combinator, successfully getting into its Spring 2024 cohort.
Now, their San Francisco-based company, Theseus, has just raised $4.3 million in seed funding in a round led by First Round Capital, with additional backing from Y Combinator and Lux Capital, it exclusively told TechCrunch.
Theseus joins a flock of other drone-related startups. There’s Skydio, which focuses on replacing Chinese drones for U.S. law enforcement and was last valued at $2.2 billion in 2023. Shield AI, which builds reconnaissance drones, recently raised at a $5.3 billion valuation. The biggest defense tech player, Anduril, launched its own small drone last year, and is reportedly in talks to raise at a $28 billion valuation.
Theseus says it doesn’t build drones, but focuses on the hardware components and software that will enable pretty much any military drone to fly unmanned without GPS. Schoeller, Theseus’ CEO, told TechCrunch the company doesn’t build targeting systems. Its software is not deciding whether a certain spot is a legitimate military target or not — the sole focus is getting a drone from point A to B.
Theseus hasn’t won any U.S. military deals yet, and hasn’t been deployed in an actual battlefield. So it’s using its fresh capital to focus on further building out its tech, hiring for three engineering roles.
However, the viral hackathon tweet did get Theseus noticed by U.S. Special Forces, which has entered into an agreement for early testing and development. Theseus says it recently went to a secret Special Forces base to test out its latest system, sending TechCrunch a photo of it in action.
Overall, starting a company with people you’ve known for under a week “generally isn’t advised,” but in Theseus’ case, it warranted the leap of faith, Schoeller wrote on LinkedIn.
Keep reading the article on Tech Crunch
Ramp is trying to get the US government as a customer after seeing a tweet from DOGE
Expense management startup Ramp is being considered for a charge card pilot program by the U.S. government’s General Services Administration, the company confirmed to TechCrunch on Thursday.
The government’s internal expense card program, dubbed SmartPay, is a $700 billion program. It is estimated that the charge card pilot program contract for which Ramp is being considered is worth up to $25 million, according to a report by Pro Publica.
Pro Publica claims that fintech Ramp has been lobbying for the administration’s attention since January, before President Trump was sworn in.
In January, Ramp co-founder CEO Eric Glyman and Ramp VC investor Kyle Harrison wrote a blog post titled “The Efficiency Formula” in which they listed the ways they imagined the government could “eliminate inefficient spending.” Harrison is a general partner at the firm Contrary.
The post seemed to be an appeal to Elon Musk’s government agenda — which would be formally created a few days later as the Department of Government Efficiency — considering Ramp has ties to Musk’s and Trump’s world. Ramp’s investors include Peter Thiel’s Founders Fund; Keith Rabois of Khosla Ventures; Thrive Capital, which was founded by Joshua Kushner, brother of Trump’s son-in-law Jared; Trump ally 8VC’s Joe Lonsdale and Jeb Bush, former governor of Florida and brother of former Republican President George W. Bush.
Ramp “is competing in a standard procurement process for a SmartPay pilot program based on the strength of our solution,” Lindsay McKinley, head of communications told TechCrunch on Thursday.
She added: “Ramp’s technology has prevented billions of dollars in wasted spend across the economy, and if chosen, we’ll bring those same results to the American taxpayer.”
Despite McKinley’s strong rhetoric, she’s referring to how Ramp positions itself as a money-saving option for corporations. It offers similar spend management features as other corporate expense management platforms, like setting parameters to identify expenses that don’t conform to policies. The federal government has many such policies for employees in place.
McKinley said that the startup saw a public post on X shared by the Department of Government Efficiency, better known as DOGE, on February 18 that said “the US government currently has ~4.6M active credit cards/accounts, which processed ~90M unique transactions for ~$40B of spend in FY24.”
A former customer, Ramp claims, introduced Ramp to GSA a few days later.
“Since then we have demonstrated the product and are now part of a standard RFI process,” she said. “We have no indication of whether we’ll be selected.”
In March, Ramp doubled its valuation to $13 billion after a $150 million secondary share sale. The startup has raised over $1 billion in equity financing and $700 million in committed debt funding since its 2019 inception.
Keep reading the article on Tech Crunch
OpenAI launches Flex processing for cheaper, slower AI tasks
In a bid to more aggressively compete with rival AI companies like Google, OpenAI is launching Flex processing, an API option that provides lower AI model usage prices in exchange for slower response times and “occasional resource unavailability.”
Flex processing, which is available in beta for OpenAI’s recently released o3 and o4-mini reasoning models, is aimed at lower-priority and “non-production” tasks such as model evaluations, data enrichment, and asynchronous workloads, OpenAI says.
It reduces API costs by exactly half. For o3, Flex processing is $5/M input tokens (~750,000 words) and $20/M output tokens versus the standard $10/M input tokens and $40/M output tokens. For o4-mini, Flex brings the price down to $0.55/M input tokens and $2.20/M output tokens from $1.10/M input tokens and $4.40/M output tokens.
The launch of Flex processing comes as the price of frontier AI continues to climb, and as rivals release cheaper, more efficient budget-oriented models. On Thursday, Google rolled out Gemini 2.5 Flash, a reasoning model that matches or bests DeepSeek’s R1 in terms of performance at a lower input token cost.
In an email to customers announcing the launch of Flex pricing, OpenAI also indicated that developers in tiers 1-3 of its usage tiers hierarchy will have to complete the newly introduced ID verification process to access o3. (Tiers are determined by the amount of money spent on OpenAI services.) O3’s reasoning summaries and streaming API support are also gated behind verification.
OpenAI previously said ID verification is intended to stop bad actors from violating its usage policies.
Keep reading the article on Tech Crunch
Carbon removal startup Holocene bought by oil and gas giant Occidental
Occidental has bought Holocene, marking the second direct air capture startup the fossil fuel company has bought in two years.
The deal was executed through Oxy Low Carbon Ventures, a subsidiary of the oil and gas company, for an undisclosed amount. HeatMap first reported the news.
Holocene had been racing to advance its amino acid-based carbon removal technology following a $10 million deal it signed in September with Google to deliver 100,000 metric tons of carbon removal by the early 2030s.
At $100 per metric ton, the price was significantly lower than what competitors could offer today. Currently, removing carbon dioxide directly from the atmosphere is estimated to cost around $600 per metric ton.
Occidental’s interest in carbon capture stems from a technique known as enhanced oil recovery, in which CO2 is injected underground to stimulate oil wells. The company bought another direct air capture startup, Carbon Engineering, in 2023 for $1.1 billion.
An Occidental spokesperson told HeatMap that the company will be using Holocene’s technology to further its direct air capture research and development.
Direct air capture qualifies for tax credits under the Inflation Reduction Act, with the final incentive dependent on whether the equipment uses zero-emission power and if the captured carbon dioxide is used for enhanced oil recovery.
Keep reading the article on Tech Crunch
As the trade war escalates, Hence launches an AI ‘advisor’ to help companies manage risk
President Donald Trump’s tariffs have underscored the increasing geopolitical risk that almost all businesses now face. As the situation continues to shift with Trump’s unpredictable deal-making, it’s also becoming clear how challenging it is for companies, nonprofits, consultants, and lawyers to keep up with the rapid day-to-day changes.
“We are drowning in trade updates every hour of every day,” Matthew Oresman, London managing partner of Pillsbury Winthrop Shaw Pittman, a global law firm, told TechCrunch.
The firm, whose clients span multinationals and high-net-worth individuals, as well as companies in tech, energy, and AI, is one of the first customers for London-based startup Hence AI’s new software product, Hence Global. The product uses AI to help organizations monitor geopolitical and business risk.
The tool does two things. First, it helps companies of any size track risk, and it advises them on actions they can take to mitigate that risk. Second, it helps service providers, like consulting and law firms, generate meaningful analysis about the world for their clients.
Sean West, CEO and co-founder of Hence AI (formerly Hence Technologies), said to think of Hence Global as “an AI-powered business advisor that’s riding alongside you.” At only $1,500 per year for the base product, Hence is far more affordable than your typical consultant, argued West.
“We want to democratize access to this information,” West told TechCrunch in an exclusive interview, noting that organizations like startups and NGOs can’t afford to call their lawyers every time they have a question about exposure or spend half a million dollars on a consulting contract.
“Big companies know how to buy expensive advice, and the richest people will always talk to the smartest people, and they’ll pay for them,” West continued. “But most of the market can actually be served by technology.”
West previously served as global deputy CEO of geopolitical advisory firm Eurasia Group and recently published a book, “Unruly: Fighting Back when Politics, AI, and Law Upend the Rules of Business.” One of his co-founders, Steve Heitkamp, is a Palantir alum with a background in political risk and counterterrorism.
Hence Global is built on Palantir’s Foundry and Artificial Intelligence Platform, which allows the startup to blend different AI models to understand, summarize, and analyze relevant information based on a customer’s specific needs and industry. The system pulls in data from news headlines (and only the headlines, with a link to the source, says West), Wikipedia, Securities and Exchange Commission filings, press releases, and other public data like sanctions lists or World Bank information.
West walked me through how the product works to show its potential impact. Customers start by creating a persona. For the purposes of the demo, we told the system that we were a cross-border cryptocurrency infrastructure company that offers stablecoin payments, crypto custody, and regulatory intelligence. We also told Hence Global that we wanted “a continuous, forward-looking analysis of geopolitical developments that could impact our operating environment,” and gave it a handful of topics to track.
The idea is that, every day, Hence Global will generate a daily update with relevant news and information. In the case of the fictional crypto company, one of the stories the software flagged was that Trump’s trade war caused market volatility that resulted in Bitcoin and other crypto stocks falling.
“Basically, it does the work that a mid-level analyst would have done in my organization,” West said. He added that it would’ve taken said analyst all day to produce a memo that Hence Global churns out in a minute.
Hence Global’s daily briefing is great for companies that want to understand their own geopolitical and business risk, but it’s also helpful for services companies that are tracking this information for their clients.
Crucially, when a customer asks Hence Global to monitor a company, the system asks if that company is a client, a competitor, or a supplier, which will cause the platform to think about how it provides information and analyses differently.
“We were desperate for these kinds of tools that can synthesize [information], help us write the client alerts, give us that big kernel of information that we can augment and put our legal knowledge on,” Oresman said. “There’s just a fire hose of information out there. Having something like this actually gets it to a water fountain so we can actually do something useful with it.”
Hence Global hasn’t been live for more than a couple of months, but Trump’s tariff situation has already begun bringing in new clients on top of existing customers like TravelPerk, Diversifi Capital, and Three Crowns.
Rohitesh Dhawan, CEO of the International Council on Metals and Mining, told TechCrunch he uses Hence Global to monitor market sentiment and policy.
“We try to show the world that it is possible to produce mining in a responsible way,” Dhawan said. “But to do that, you’ve got to be really well connected to the pulse of society and what people care about and the issues that are top of mind, and that’s why we turn to Hence as a way to help us do that, because things are just moving so quickly in the world in general.”
Dhawan likened Hence Global to Uber Eats — a product you didn’t know you needed until it showed up and made your life easier. He said that businesses in resource-based industries, like agriculture or oil and gas, and companies that are highly regulated or highly exposed to public sentiment, such as tech startups, would benefit most from using Hence Global.
“This was the kind of always-on monitoring that we just weren’t doing before,” Dhawan said. “What Hence is helping me do as a CEO is to quickly get to the, ‘So what?’”
Hence Global is the startup’s second product after Hence Legal, a “solution for outside counsel management, matter management, and litigation management,” per West. Hence AI, whose team is spread around the U.K., Rwanda, the U.S., and the Netherlands, has raised about $5.2 million to date and is actively looking to raise more for this product from investors who align with its mission.
When asked about whether Hence AI would be open to being acquired by larger research houses, consultancies, or law firms, West said he and his team are less interested in going in-house at any one organization.
“We’re trying to do big things,” he added. “And I think the big opportunity here is tapping the market of people who are global, but can’t access advice. And I don’t think that necessarily is what a large corporate service provider would do with this.”
Keep reading the article on Tech Crunch
Chinese shopping app Taobao joins DHgate in Top 5 on US App Store
The Chinese e-commerce marketplace app DHgate, which is now the No. 2 free iPhone app in the U.S., isn’t the only one that’s oddly benefiting from President Trump’s tariffs on U.S. imports from China. Another Chinese shopping app, Taobao, has now also entered the Top 5 as of Thursday.
U.S. consumers began flocking to these apps over the past several days in the wake of numerous TikTok videos from Chinese manufacturers explaining how much of the luxury goods market operates out of China. The videos claim that many products from top luxury brands — like clothing, handbags, shoes, and accessories — are actually originally made in China. The items are then shipped over to the brand’s home country, like Italy or France, where they’re repackaged after the brand’s label is applied, according to these videos.
Other U.S. and Chinese TikTok creators then pointed to e-commerce apps like DHGate and Taobao as a way to buy directly from the Chinese manufacturers, forgoing the huge markup the luxury brands charge.
Already wary of the increasing prices on popular apps like Shein and Temu, U.S. consumers quickly began downloading these alternatives.
In April, Taobao’s estimated downloads totaled approximately 185,000 — a 514% increase from the 30,000 it saw during the same period last month, according to new data from app intelligence provider Appfigures. As with DHgate, Taobao’s downloads surged over this past weekend, with installs increasing 5.7x between last Friday and Saturday, the firm said.
On Saturday, Taobao entered the Top Overall ranks on the iOS App Store (excluding games) at No. 461. By Sunday, it was No. 188. And by Thursday, it reached No. 5.
The change in rank represents the growth in app installs as well as their velocity and other factors.
Taobao also grew from the No. 49 Shopping app on Saturday to become the No. 2 app, topping Walmart, Amazon, Shein, and Temu. (Chinese app Alibaba.com is also climbing the charts here, now the No. 6 shopping app on the U.S. App Store.)
Notably, Appfigures says that Taobao has never been in the Top Overall charts on the U.S. App Store according to its data, which goes back to January 1, 2017.
While switching shopping apps won’t actually save U.S. consumers from tariffs on Chinese imports, shoppers likely think buying direct from manufacturers could potentially lower the overall cost of their purchases. For other consumers, it’s simply a way to seek out luxury-style goods or dupes at a better price.
Of course, it’s still a case of “buyer beware” on any of these marketplaces, where quality can be hit or miss. Users are encouraged to read the individual sellers’ reviews and view other buyers’ photos of the items before making purchases.
Keep reading the article on Tech Crunch
Former Y Combinator president Geoff Ralston launches new AI ‘safety’ fund
Geoff Ralston, well-known in the startup community for his years at Y Combinator, is back in the formal investing ring, he announced Thursday.
His new fund is called Safe Artificial Intelligence Fund, or SAIF, which is both an explanation of its thesis and a play on words.
Ralston is specifically looking for startups that “enhance AI safety, security, and responsible deployment,” as his fund’s website describes. He plans to write $100,000 checks as a SAFE, “pun intended,” he says, with a $10 million cap. A SAFE is, of course, the invest now/price later pre-seed investment tool pioneered by Y Combinator (it stands for simple agreement for future equity).
While most VCs these days are looking to invest in AI startups, Ralston’s take is a bit more focused on the idea of safe AI, even though he admits the concept is a bit broad.
“The vast majority of AI projects out in the world today are using the technology to solve problems or create efficiencies or create new capabilities. They are not necessarily intrinsically unsafe, but safety is not their primary concern,” Ralston tells TechCrunch. “I intend to fund startups whose primary objective is safe AI — as I have (very broadly) defined it.”
That list includes startups focused on improving the safety of AI, like those that clarify an AI’s decision-making process or benchmark AI safety. It includes products that protect intellectual property, those that ensure an AI conforms to compliance requirements, fight disinformation, and detect AI-generated attacks. He also wants to invest in functional AI tools with built-in safety in mind, such as better AI forecasting tools and AI-enabled business negotiation tools that won’t reveal corporate secrets to outsiders.
This might sound like a list of AI startups that many VCs are pursuing, but there are areas Ralston says he won’t back. One example is fully autonomous weapons.
“There are certainly uses of AI which would (will) be unsafe: using the technology to create bioweapons, to manage conventional weapons without a human in the loop, etc.,” he explained.
In fact, he’d like to fund “weapon safety systems” that could detect or prevent attacks from AI weapons.
This is an interesting contrarian viewpoint from many of today’s defense tech founders and VCs. As TechCrunch has previously reported, some of the people building AI weapons have increasingly been floating the idea that such weapons would be better operating without a human.
Still, all things AI is a crowded field for VCs these days. That’s where Ralston hopes his YC connections could give him an advantage. Ralston departed YC in 2022, after three years as president (succeeded by Garry Tan) and over a decade as an adviser.
Ralston plans to offer mentoring of the kind he did at the storied startup accelerator and has promised to coach them through how to apply to YC. And he’s offering to help them tap into his considerable investor network.
Ralston declined to say how big this fund is, how many startups he intends to back, or who his LP backers are.
Keep reading the article on Tech Crunch
Google’s latest AI model report lacks key safety details, experts say
On Thursday, weeks after launching its most powerful AI model yet, Gemini 2.5 Pro, Google published a technical report showing the results of its internal safety evaluations. However, the report is light on the details, experts say, making it difficult to determine which risks the model might pose.
Technical reports provide useful — and unflattering, at times — info that companies don’t always widely advertise about their AI. By and large, the AI community sees these reports as good-faith efforts to support independent research and safety evaluations.
Google takes a different safety reporting approach than some of its AI rivals, publishing technical reports only once it considers a model to have graduated from the “experimental” stage. The company also doesn’t include findings from all of its “dangerous capability” evaluations in these write-ups; it reserves those for a separate audit.
Several experts TechCrunch spoke with were still disappointed by the sparsity of the Gemini 2.5 Pro report, however, which they noted doesn’t mention Google’s Frontier Safety Framework (FSF). Google introduced the FSF last year in what it described as an effort to identify future AI capabilities that could cause “severe harm.”
“This [report] is very sparse, contains minimal information, and came out weeks after the model was already made available to the public,” Peter Wildeford, co-founder of the Institute for AI Policy and Strategy, told TechCrunch. “It’s impossible to verify if Google is living up to its public commitments and thus impossible to assess the safety and security of their models.”
Thomas Woodside, co-founder of the Secure AI Project, said that while he’s glad Google released a report for Gemini 2.5 Pro, he’s not convinced of the company’s commitment to delivering timely supplemental safety evaluations. Woodside pointed out that the last time Google published the results of dangerous capability tests was in June 2024 — for a model announced in February that same year.
Not inspiring much confidence, Google hasn’t made available a report for Gemini 2.5 Flash, a smaller, more efficient model the company announced last week. A spokesperson told TechCrunch a report for Flash is “coming soon.”
“I hope this is a promise from Google to start publishing more frequent updates,” Woodside told TechCrunch. “Those updates should include the results of evaluations for models that haven’t been publicly deployed yet, since those models could also pose serious risks.”
Google may have been one of the first AI labs to propose standardized reports for models, but it’s not the only one that’s been accused of underdelivering on transparency lately. Meta released a similarly skimpy safety evaluation of its new Llama 4 open models, and OpenAI opted not to publish any report for its GPT-4.1 series.
Hanging over Google’s head are assurances the tech giant made to regulators to maintain a high standard of AI safety testing and reporting. Two years ago, Google told the U.S. government it would publish safety reports for all “significant” public AI models “within scope.” The company followed up that promise with similar commitments to other countries, pledging to “provide public transparency” around AI products.
Kevin Bankston, a senior adviser on AI governance at the Center for Democracy and Technology, called the trend of sporadic and vague reports a “race to the bottom” on AI safety.
“Combined with reports that competing labs like OpenAI have shaved their safety testing time before release from months to days, this meager documentation for Google’s top AI model tells a troubling story of a race to the bottom on AI safety and transparency as companies rush their models to market,” he told TechCrunch.
Google has said in statements that, while not detailed in its technical reports, it conducts safety testing and “adversarial red teaming” for models ahead of release.
Keep reading the article on Tech Crunch
Automakers selling cars in China banned from using ‘autonomous driving’ in ads
China is cracking down on how automakers advertise driver assistance features, banning terms like “autonomous driving,” “self-driving,” and “smart driving,” Reuters reported, citing a transcript of a meeting between the government and industry representatives.
The updated rule will also prohibit automakers from rolling out improvements via software updates to advanced driving assistance systems in vehicles already in the hands of customers. Automakers must now test any upgrades or new features in ADAS and receive government approval before rolling it out via software update.
Wireless software updates that improve or fix features in vehicles — and first popularized by Tesla — has become a critical capability for automakers trying to stay competitive.
The mandate stems from growing concerns over vehicle safety and incidents of fatal crashes involving advanced driver assistance systems. Last month, a Xiaomi Su7 sedan caught fire after crashing into a pole just seconds after the driver took over control from the ADAS.
It also follows Tesla’s rollout of its “Full Self-Driving software” branding used for its advanced driver assistance system. FSD is not a self-driving system and requires a human driver to remain engaged. In China, Tesla changed the FSD name to “Intelligent Assisted Driving.”
Keep reading the article on Tech Crunch
AI benchmarking platform Chatbot Arena forms a new company
Chatbot Arena, the crowdsourced benchmarking project major AI labs rely on to test and market their AI models, is forming a company called Arena Intelligence Inc., reports Bloomberg.
In a blog post published Thursday, Chatbot Arena said that the company will “give [it] the resources to improve [its platform] significantly over what it is today.” The team also pledged to continue to provide neutral testing grounds for AI not influenced by outside interests.
Founded in 2023, Chatbot Arena has become something of an AI industry obsession. Primarily run by UC Berkeley-affiliated researchers, Chatbot Arena has partnered with companies such as OpenAI, Google, and Anthropic to make flagship models available for its community to evaluate.
Chatbot Arena was previously funded through a combination of grants and donations, including from Google’s Kaggle data science platform, Andreessen Horowitz, and Together AI. The organization’s fledgling company hasn’t disclosed any potential new backers yet — nor has it decided on a business model.
Keep reading the article on Tech Crunch
Florida draft law mandating encryption backdoors for social media accounts billed ‘dangerous and dumb’
A Florida draft bill that would require social media companies to provide encryption backdoors for law enforcement officials to access user accounts has cleared a key legislative hurdle and will now advance to the state’s Senate floor for a vote.
Florida lawmakers unanimously approved pushing the bill through committee, per Florida Politics.
The “Social Media Use by Minors” (SB 868) bill, if passed into law, would require “social media platforms to provide a mechanism to decrypt end-to-end encryption when law enforcement obtains a subpoena.” The bill would also require social media companies to allow parents or guardians access to a child’s account, and would prohibit child accounts from using features that allow the use of disappearing messaging, the bill reads.
Critics, including the tech companies and industry organizations that oppose the bill, have long argued that weakening encryption would make people less safe by compromising the security of their private messages, and could result in data breaches.
In a blog post last week, the digital rights group Electronic Frontier Foundation criticized the bill, arguing that encryption is the “best tool we have to protect our communications online,” and that passing the law would likely result in companies removing encryption for minors and making those users less safe.
“The idea that Florida can ‘protect’ minors by making them less safe is dangerous and dumb,” wrote the EFF.
The Florida bill builds on a state law passed last year restricting social media for people under the age of 16. The law remains largely on hold while it remains under scrutiny in the courts amid questions about the law’s constitutionality.
Tech companies, like Apple, Google, and Meta, are increasingly end-to-end encrypting their users’ data so that their private content is only accessible to the user, not even the companies themselves. This also helps to protect users’ private messages from hackers or malicious company insiders. By encrypting user data, the tech companies say they also cannot provide law enforcement with information that they themselves cannot access.
It’s not clear if the proposed Florida bill, as written, would require social media companies to comply with only a subpoena, which are typically issued by law enforcement agencies and without judicial oversight.
Subpoenas are usually not signed by a judge but can still be used by law enforcement to compel limited amounts of account information, such as names, email addresses, or phone numbers, from tech companies about their users. Companies will often demand to see a court-authorized search warrant, which requires police to present a court with a higher degree of evidence of suspected criminality, before turning over a user’s private messages.
A corresponding bill going through the Florida House (HB 743) has a final committee vote to clear before it will proceed to the House floor for a vote, per Florida Politics.
Keep reading the article on Tech Crunch
Instagram’s new Blend feature creates a custom reels feed for you and your friends
Instagram on Thursday announced that it’s rolling out Blend, a new feature that lets you create a custom, personalized reels feed for you and your friends. Blends are invite-only and can be created with a singular friend or with a group chat. The custom feeds are refreshed with new content each day.
The launch doesn’t come as a surprise, as TechCrunch reported in March 2024 that the social network was developing the feature.
Once you join or accept a Blend invite, you will be able to browse through reels picked for each person in the chat. When someone in the group reacts to a reel, Instagram will notify you so you can keep a convo going in the DM (direct messaging) chat.
The idea behind the feature is to explore what sorts of reels your friends are into, while also connecting with them and discovering new content together. Blend aims to introduce a new social element to Instagram, bringing it closer to its roots of sharing moments with friends, something that’s been overshadowed in recent years by an influx of influencers and ads.
With the launch of Blend, Instagram is giving users access to a feature that isn’t available on TikTok, which is one of Instagram’s biggest rivals.
For Instagram, Blend could be a way to boost reels’ discovery and watch time on the social network, as it’s encouraging users to watch short-form content together.
To create a Blend, you need to open a one-on-one or group DM chat and tap the new Blend icon at the top of the chat. Then, you need to select “Invite” to invite people from your DM chat to join the Blend. The Blend will be created when at least one member of the chat accepts your invite. You can revisit a Blend by going to the DM chat and tapping the Blend icon.
It’s worth noting that Blend is somewhat similar to Spotify’s functionality of the same name, which allows people to combine their tastes into one shared playlist that refreshes daily.
Keep reading the article on Tech Crunch
Here are all the tech companies rolling back DEI or still committed to it — so far
Companies around America have started cutting DEI programs and eliminating DEI commitments from public documents in response to legal and political threats from the Trump administration.
Just a few weeks ago, Attorney General Pam Bondi instructed the Department of Justice to “investigate, eliminate, and penalize,” DEI programs deemed illegal in private sector companies that receive federal funds.
Trump has signed an executive order banning DEI in the federal government with the approval of his right-hand companion Elon Musk, an immigrant from South Africa and a long-time denouncer of DEI.
The reaction to the government’s anti-diversity push has taken many forms. One contractor said NASA has started to scrub mention of terms like “environmental justice” or “anything specifically targeting women.” PBS had to shutter its DEI office.
In corporate America, McDonald’s, Harley-Davidson, Booz Allen, John Deere, Tractor Supply Company, Polaris, Lowe’s, Ford, Molson Coors, Walmart, Nissan, Accenture, and Target, among others, have all announced some form of rollback to their DEI policies. Goldman Sachs said in February that it would ditch its requirement for corporate boards filing to go public to include women and people of color.
Other companies are trying to strike a balance. Some companies like Apple, Costco, Delta, McKinsey, and JPMorgan have indicated a commitment to sticking with their diversity efforts.
Last year, companies like Telsa, GameStop, Workday, and Salesforce started scrubbing mentions of DEI from their 10-K filings, a trend that has accelerated so far this year. Here’s a list of where DEI stands, so far, among tech companies in 2025.
We’ve reached out to each of the companies for comment.
Amazon
Amazon’s website “positions” page still showcases its diversity, equity, and inclusion commitment. However, it is winding down some DEI “outdated programs and materials,” according to a leaked memo reported by Bloomberg. It removed a post from 2021 that expressed a focus on DEI and removed some mentions of diversity from its latest annual 10-K report.
In a statement to TechCrunch, an Amazon spokesperson pointed to the “positions” page saying “as stated here, we’re committed to creating a diverse and inclusive company.”
Apple
Apple’s website still says, as it has for years, “We’re continuing to create a culture of inclusion, increasing representation across teams, and holding ourselves accountable at every level.”
Shareholders at Apple’s latest meeting rejected a proposal from a conservative group to strike DEI policies from the companies. Apple previously advised its shareholders to vote against the proposal.
Google is eliminating diversity hiring targets and evaluating releasing new diversity reports, which it has done since 2014, the Wall Street Journal reported. It also said it would review its DEI initiatives to ensure all comply with the latest executive orders aimed at curbing DEI.
Google also removed underrepresented language from its grant website and scrubbed mention of diversity and equity from its responsible AI team webpage. Furthermore, Google updated its 10-K filing, making no mention of diversity in its latest one, compared to the eight times it was mentioned in its 2023 filing. Late last year, it removed many cultural events from the default Google Calendar.
IBM
IBM still has its inclusion page up. Working with a shareholder, the conservative think tank Heritage Foundation requested the company drop its DEI pay incentive targets, though IBM has moved to block the proposal.
Intel
Intel cut back some of the language about DEI on its recent 10-K filing and removed some diversity targets. On its website, though, the company still states that “diversity and inclusion are among the most important forces driving the company’s evolution and reinvention.”
Medium
The publishing platform Medium said in a blog post that it would remain committed to DEI. The company’s CEO said most people value understanding and respect over hate and division.”
Meta is eliminating some DEI programs targeting hiring, suppliers, and training according to a leaked internal memo reported by Axios. Meta also said it eliminated its DEI team and ended representation goals to eliminate any impression “that decisions are being made based on race or gender,” the memo read. “While this has never been our practice, we want to eliminate any impression of it.” Meta declined to comment.
Microsoft
Microsoft released its 2024 inclusion report in October where CEO Satya Nadella reiterated the importance of a diverse workforce. However, in July of that year Microsoft laid off its internal DEI team because of “changing business needs,” according to a memo leaked to Business Insider.
Nvidia
Nvidia still has its “diversity, inclusion, and belonging” page up and released its 2024 Sustainability Report where it has a section still dedicated to “People, Diversity and Inclusion.”
OpenAI
OpenAI quietly changed its “Commitment to Diversity” website to now read “Building Dynamic Teams,” and removed all mention of diversity and inclusion from the web page.
Oracle
Oracle still has its “Culture and Inclusion” page up, which reads that “diverse perspectives make our teams stronger and empower collaboration.
Salesforce
Salesforce is getting rid of its diversity hiring goals. It also removed mention of “diversity” from its 10-K filing.
Tesla
Tesla scrubbed mention of diversity, equity, and inclusion from its 10-K last year. Elon Musk, the CEO of the company, is an outspoken critic of DEI. The company only released one diversity, equity, and inclusion report, back in 2020.
Workday
Although Workday erased mention of diversity targets in its 2023 10-K form after expressing a commitment to increase the representation of Blacks and Latinos at the company, its DEI page is still up on the website. On that page, the company wrote that “diversity isn’t just a business imperative. It’s core to everything we do.” It released a 2024 Global Impact Report and it still has a chief diversity officer.
Uber
Uber eliminated an incentive tied to determining executive compensation pay by measuring the diversity progress of the company; it got rid of the promise to consider women and people of color candidates when looking to fill empty board seats or positions; and it removed a section in its proxy statement that highlights diversity as a value. It removed the entire diversity and inclusion section from its latest 10-K filing and made no use of the word diversity at all, according to the SF Examiner.
UnitedHealth Group
UnitedHealth Group removed mention of DEI from much of its website. For example, several pages once dedicated to DEI no longer load, and the company also removed a 2022 blog post with a vice president about DEI.
Yahoo
Yahoo removed several pages and sections that relate to or mention diversity, equity, and inclusion. For example, its DEI page now redirects to the company’s executive leadership page, and the 2022 diversity report no longer loads. (Disclosure: Yahoo is the parent company of TechCrunch. On March 21, Regent announced it would acquire TechCrunch.)
Zoom
Zoom laid off its DEI team during a round of layoffs last year, telling staff it would work with external consultants who will engage “all of our employees,” while focusing on “inclusion,” according to a memo leaked to Bloomberg. It doesn’t appear to have released a diversity report since 2022, the first of which came out in 2020, according to its website.
Keep reading the article on Tech Crunch
OpenAI’s Stargate project sets its sights on international expansion
Stargate, a $500 billion project headed up by OpenAI, Oracle, and SoftBank to build AI data centers and other AI infrastructure in the U.S., is considering investments in the U.K. and elsewhere overseas, according to a Financial Times report.
While Stargate was initially launched as a way to boost U.S. AI infrastructure, the project is allegedly weighing international expansion. In addition to the U.K., Germany and France are on the table, per the Financial Times’ reporting.
Stargate remains focused on the U.S. at the moment, to be clear, as originally pitched — and it’s still in the process of raising its first $100 billion. SoftBank is expected to put forward tens of billions of dollars as a mixture of debt and equity.
When Stargate was announced in January, President Donald Trump praised the initiative as a “declaration of confidence in America.”
Keep reading the article on Tech Crunch