📰 TechCrunch · 90
United StatesA startup claims it broke through a bottleneck that’s holding back LLMs
A Miami-based AI startup named Subquadratic has claimed to have overcome a major mathematical bottleneck limiting large language models (LLMs). The company introduced a new model called SubQ, which it says is faster, cheaper, and more energy-efficient than existing models. SubQ is reported to handle significantly more text at once and perform well on key tasks compared to leading models from companies like Google DeepMind, OpenAI, and Anthropic. However, initial skepticism existed due to limited evidence, though the company has since shared results from an independent evaluation.
Anthropic becomes first AI startup to join the Frontier carbon removal coalition
Anthropic has joined the Frontier carbon removal coalition, contributing to a new $915 million funding round. This makes Anthropic the first AI startup to join the group. The funding brings Frontier's total commitments to $1.8 billion. Frontier has already allocated nearly $700 million to over 50 projects aimed at removing 1.8 million tons of carbon. Anthropic's participation marks its first climate-related commitment, though the company has not released a sustainability report and prefers an 'all of the above' energy strategy, which includes purchasing polluting power.
The US says ASML’s top chip tool may be in China. ASML says it isn’t
The U.S. Commerce Secretary has raised concerns with ASML executives about potential unauthorized shipments of extreme ultraviolet (EUV) lithography machines to China, which would violate existing export controls. ASML denies these claims, stating no such machines exist in China. The U.S. government has reportedly seen evidence of related components being sent to China but has not provided proof to either Bloomberg or ASML.
Fake MAGA? Anti-AI Group’s Conservative Branding Crumbles
An organization called Humans First, which presents itself as a conservative group opposing the expansion of AI data centers, has been revealed to have ties to leftist activists. The group's leader, Amy Kremer, a former Tea Party figure associated with the January 6 events, frames the movement as a grassroots conservative response to perceived elite control by Silicon Valley. However, investigations show that key members of Humans First have backgrounds in leftist activism, including work with the Sunrise Movement, a group known for opposing Trump and advocating for aggressive climate action.
Telegram ban in India sparks a rush to VPNs, rival apps
India temporarily banned access to the messaging app Telegram due to concerns about exam-related fraud. In response, users significantly increased their downloads of virtual private network (VPN) apps, with some seeing download spikes of over 100%. This surge in demand led to an increase in the rankings of several VPN services on Indian app stores.
Source: Elastic agrees to buy CRV-backed DeductiveAI for up to $85M
DeductiveAI, a startup using AI to detect and fix software bugs, has agreed to be acquired by Elastic for up to $85 million. The startup raised $7.5 million in a seed round led by CRV in late 2023. Elastic and Deductive have not responded to requests for comment. The acquisition highlights a growing trend where established tech companies are acquiring AI-native startups to incorporate agentic technologies into their products.
UK unveils sweeping social media ban for users under 16
The U.K. government, led by Prime Minister Keir Starmer, plans to implement a ban on social media use for individuals under 16 years old. This measure applies to major platforms such as Snapchat, TikTok, YouTube, Instagram, Facebook, and X. Messaging apps like WhatsApp and Signal will remain unaffected. Additionally, AI-based 'romantic companion' chatbots will be restricted to users aged 18 and older. The U.K. claims this regulation will be more extensive than any previously enacted by other nations. Other countries, including Australia, Canada, France, and Denmark, are also exploring similar
AI inference startup Baseten reportedly raising $1.5B months after its last mega-round
AI inference company Baseten is reportedly nearing completion of a $1.5 billion funding round at a $13 billion valuation, according to the Wall Street Journal. This follows a $300 million Series E round at a $5 billion valuation just five months earlier and a $150 million Series D round nine months prior. The new round is described as a 'split-priced' deal, with some investors participating at the higher valuation and others at a lower one. Baseten, founded in 2019, operates in the growing 'inference gold rush,' where venture capital firms are investing heavily in companies focused on the AI '
Trump's fight with Anthropic is now a fight over cybersecurity
The Trump administration's restrictions on Anthropic's AI models have raised concerns among cybersecurity experts about potential negative impacts on U.S. cyber defenses. Prominent security leaders, including former Facebook security chief Alex Stamos, have criticized the administration's actions, warning that they could discourage American AI companies from developing tools that help identify and fix vulnerabilities. An open letter signed by nearly 150 security professionals called for the reversal of these restrictions. The dispute arose following Anthropic's release of its first Mythos-5 AI
Snap spins off AI video team into new company, Dotmo, due to costs
Snap Inc. has decided to spin off its internal generative AI video team into a new company called Dotmo. This decision was primarily driven by the high costs associated with maintaining such operations within Snap. Despite being a separate entity, Dotmo will maintain strong ties with Snap through a technology licensing agreement and the involvement of key personnel, including Snap's Chief Technology Officer, Bobby Murphy, who will serve as a lead investor in Dotmo. Snap will hold a significant equity stake in Dotmo, which could yield substantial returns if the new company succeeds.
OpenAI is bringing on some big guns in the lead-up to its IPO
OpenAI has hired Noam Shazeer, a prominent figure from Google DeepMind and co-founder of Character AI, and Dean Ball, a former Trump administration official involved in AI policy. Shazeer previously worked at Google and was part of the development of the Transformer architecture. His past involvement in internal discussions on sensitive topics has raised questions about potential controversies. OpenAI is also strengthening its policy expertise with Ball's addition.
AI is hurting Apple in more ways than one: it may force iPhone price increases
Apple CEO Tim Cook has warned that rising costs of memory and storage chips due to high demand for AI technology could lead to unavoidable price increases for upcoming products such as the iPhone, Mac, iPad, and others. These cost increases have surged fourfold since last year, according to Cook, who described the situation as 'unsustainable.' Experts suggest the iPhone is likely to see price hikes, with the next model expected to launch in September. The impact of these rising costs on Apple's future business results has already been acknowledged by both outgoing CEO Tim Cook and incoming CEO
Rivian owners file lawsuit alleging false promises on self-driving features
Rivian owners have filed a class-action lawsuit against the electric vehicle manufacturer, alleging that the company made false claims about the autonomous driving capabilities of its R1T and R1S models. The lawsuit states that Rivian misrepresented the functionality of its Driver+ system, suggesting it could enable hands-free, eyes-off driving under certain conditions, which aligns with Level 3 autonomy standards. The plaintiffs claim this was part of a coordinated marketing campaign over a five-year period.
Almost half of U.S. singles feel negatively about AI in dating, Match says
Match Group, owner of dating apps such as Tinder, Hinge, and OkCupid, conducted a survey indicating that nearly half of U.S. singles aged 18 to 39 hold a negative view of AI's role in dating. The survey found that 47% of respondents disapprove of AI's use in romantic contexts, with higher opposition among younger women. While many reject the idea of dating AI companions, a majority still see potential benefits in using AI features within dating apps.
Amazon hopes to challenge Nvidia more directly by selling its AI chips
Amazon Web Services (AWS) is considering selling its custom-built AI chips, called Trainium, to external companies for use in data centers. This move could position AWS as a significant competitor to Nvidia in the AI chip market. According to AWS CEO Andy Jassy, if AWS were a standalone chip manufacturer, its annual revenue could reach approximately $50 billion. However, AWS has previously avoided selling its chips due to various internal considerations.
Waymo recalls nearly 4,000 robotaxis to stop them driving into highway construction zones
Waymo has recalled nearly 4,000 robotaxis to prevent them from driving on highways while addressing issues with their ability to navigate highway construction zones. At least 13 incidents were reported where robotaxis entered closed construction areas, including six in Phoenix, Arizona, in April and seven in San Francisco, California, in May. Waymo has restricted robotaxis from highways since May 19 but continues to operate on surface streets. The company stated it is working on a solution and has filed a voluntary software recall with the National Highway Traffic Safety Administration (NHTSA)
SpaceX shoots 19% higher in first day of trading
SpaceX shares rose 11% in initial trades, giving the company a market capitalization of $1.97 trillion and placing it among the top 10 largest U.S. companies. The stock was priced at $135 before trading began, and if included in the S&P 500, it would represent more than 5% of the index. Elon Musk expressed skepticism about SpaceX’s chances of success prior to the IPO. Analysts note that SpaceX’s valuation is based on sales rather than profits, raising concerns about whether it aligns with fundamental business metrics. However, similar high valuations are common among top-performing S&P 500 equ
Mivo’s new app takes a mindful approach to managing screen time
Mivo Scrolling is a new app designed to help users manage their screen time through a mindful, non-shaming approach. Unlike traditional screen time apps that enforce strict limits or block apps, Mivo allows users to set customizable time limits for specific platforms and encourages reflection on their reasons for using social media. The app promotes intentional use of technology and aligns with the slowtech movement.
AI data centers just got a government-mandated fast lane to the grid
The Federal Energy Regulatory Commission (FERC) has mandated that grid operators prioritize interconnection requests from data centers and other large electricity users. Grid operators must ensure these entities can connect to the transmission system in a timely and orderly manner, with data centers covering the associated costs. FERC also encouraged the consideration of alternative transmission technologies and required grid operators to assess available generating capacity and revise electricity rates if necessary. However, the directives do not resolve the issue of insufficient generating容量
The slowtech revolution is here to kill your phone addiction and rescue your attention span
TechCrunch reports on the emergence of 'slowtech' as a movement aimed at reducing smartphone dependency and addressing issues related to attention span and digital overload. The article references Tony Fadell, co-inventor of the iPod, who reflects on the irony of seeing an old iPod Shuffle ad in a subway station filled with people using smartphones for streaming music. The piece highlights the contrast between early portable music devices and current smartphone capabilities, suggesting that while technology has advanced significantly, it has also contributed to increased screen time and device
Texas government data breach allowed hackers to steal 3 million driver’s licenses and passports
A data breach at a Texas state government department allowed hackers to access the driver’s license information and passport numbers of over 3 million individuals. The breach was reported by the state’s attorney general and involved the exposure of additional personal details such as email addresses, phone numbers, and residential addresses. The incident occurred through a third-party vendor handling hunting and fishing licenses, though the specific vendor and timeline of the breach remain undisclosed.
‘Queer Eye’s’ life coach Karamo Brown launches Kē, a wellness app featuring his AI digital clone
Karamo Brown, known for his role on 'Queer Eye,' has launched a wellness app called Kē. The app provides personalized fitness and nutrition plans, mental health resources like meditation, and a community section. A standout feature is 'AI Karamo,' which allows users to interact with a digital clone of Brown via AI.
Apple opens up App Store to new competition in Brazil
Apple has opened up its App Store to new competition in Brazil, allowing developers to distribute iOS apps through alternative app stores and process payments outside the App Store. The change is part of an agreement with Brazil's competition regulator, CADE, and follows similar moves in the EU and Japan. New protections such as a notarization process for apps distributed outside the App Store and authorization requirements for alternative app marketplaces have been introduced.
MapTap, a daily geography game, is my new Wordle
The article introduces MapTap, a new geography-based game similar to Wordle, where players identify locations on a map. The author discusses their personal experience playing the game and notes that it has become a popular alternative to Wordle among their friends.
General Intuition in talks to raise $300M at around $2B valuation
General Intuition, a New York-based startup focused on training AI agents to navigate space and time, is in discussions to raise $300 million, potentially increasing its valuation to over $2 billion. The company spun out of Medal, a video game clip-sharing platform, and has secured backing from figures like Jeff Bezos and Eric Schmidt, along with venture capital firms Khosla Ventures and General Catalyst. The startup uses Medal's extensive dataset of 2 billion annual video clips to train its AI models, emphasizing the value of first-person gameplay data for developing advanced spatial-temporal
A tech worker-backed PAC is bringing a $5M knife to Big Tech’s $100M gunfight
The Guardrails Alliance, a new super PAC backed by tech workers and labor unions, is challenging pro-tech lobbying efforts with a $5 million campaign. It supports AI legislation and targets candidates like Alex Bores, who is running against funding from groups like Leading the Future, which has over $100 million in resources.
Spotify’s reserved ticket sales to music superfans are now going live
Spotify has launched a new feature called 'Reserved by Spotify,' which identifies top fans of an artist and holds two tour tickets for them before the general public sale. The feature is initially available in the U.S. for Premium subscribers aged 18+. It will first be used by artist Role Model, with eligible fans receiving notifications today. Tickets are sold through Ticketmaster and are currently limited to Live Nation concerts. Spotify stated it does not charge fees for these transactions and plans to expand the program to include smaller venues and international users.
Roelof Botha joins SpaceX’s board of directors
Roelof Botha, former managing partner at Sequoia Capital, has joined SpaceX's board of directors following the company's historic IPO. The appointment was disclosed in an SEC filing, stating Botha will serve on the board and its audit committee. A family member of Botha's has been working at SpaceX since January 2025 in enterprise operations, earning more than $120,000 annually. Botha previously stepped down from Sequoia due to controversy involving another partner.
Adobe adds its AI assistant to Premiere, Illustrator and InDesign
Adobe is expanding the capabilities of its Firefly AI assistant, integrating it into several of its creative software applications including Premiere, Illustrator, InDesign, and Frame.io. New features include generating brand kits, product videos, and storyboards, along with saving AI-generated elements for reuse. The update also introduces tools for organizing assets, renaming clips, identifying interview questions, and checking for missing fonts within specific applications. Firefly is being enhanced with features similar to those found in Canva, such as storing AI-generated content and team
DOJ Takes Elon Musk's Side in NAACP Lawsuit Against xAI for Polluting Black Neighborhoods
The Department of Justice (DOJ) has intervened on behalf of Elon Musk's company SpaceX in a civil rights lawsuit brought by the NAACP and environmental groups against xAI, alleging violations of the Clean Air Act related to gas-burning turbines in Mississippi and Tennessee. The NAACP claims these turbines are located near residential areas, schools, and religious institutions, potentially harming public health. The DOJ has requested the court dismiss the lawsuit, arguing that the NAACP's actions threaten 'American national, economic, and energy security' by disrupting power essential for AI发展.
Pixi’s new iOS app turns text messages into interactive AR experiences
Pixi has launched an iOS app that allows users to send AI-powered AR characters through iMessage. These characters appear via the recipient's iPhone camera and can interact with their environment in real time. The app aims to enhance digital conversations by creating more immersive and spontaneous interactions compared to traditional text-based communication.
The power decisions that could shape the next century
The article discusses the impact of AI-driven power demands on the U.S. electricity system, highlighting the need for major decisions regarding infrastructure growth. It notes that data centers are consuming large amounts of electricity, comparable to entire cities, which raises questions about cost distribution, access, and grid connectivity. The piece outlines ongoing debates at key regulatory bodies like PJM and the Federal Energy Regulatory Commission, including proposals for data centers to connect directly to power plants or generate their own power. These decisions are expected to have远
NASA picks Eric Schmidt’s rocket company for Mars mission, setting up a race with SpaceX
NASA has selected Relativity Space, a rocket company acquired by former Google executive Eric Schmidt, to build a spacecraft for a Mars mission called Aeolus. This mission aims to provide the first daily, global view of Mars' atmosphere, including dust, winds, and temperatures. The spacecraft will launch in 2028 and will carry four scientific instruments. The mission follows a model similar to previous NASA contracts with companies like SpaceX and Firefly Aerospace.
How to turn off AI in your Google Docs
The article provides step-by-step instructions on how to disable the AI feature called Gemini in Google Docs. The author recounts their frustration with encountering the AI interface unexpectedly while trying to write an article and details the process of turning off the AI pop-up through the 'bottom bar preferences' option in the top menu. The author also humorously describes an attempt to ask Gemini to remove itself, which only ended the conversation without fully disabling the feature.
French president urges U.S. to share cutting-edge AI and democracies to cooperate on regulation
French President Emmanuel Macron called on wealthy democracies to collaborate on regulating advanced artificial intelligence during a high-level meeting in Evian-les-Bains, France. OpenAI CEO Sam Altman echoed this sentiment at the G7 summit, emphasizing the need for an international forum to establish AI safety guidelines. Macron praised the U.S. recognition of the risks posed by frontier AI models but criticized the Trump administration's recent restrictions on foreign access to Anthropic's AI technology as 'strictly nationalist.' The discussion took place amid a G7 working lunch featuringAI
Snap finally debuts its long-awaited AR glasses, Specs, and, oof, they aren’t cheap
Snap Inc. unveiled its consumer smart glasses, called Specs, at a spatial AI convention. Priced at $2,195, the glasses are significantly more expensive than competing products such as Meta Ray-Bans but less costly than Apple's Vision Pro. Preorders begin on June 16, with an expected release later this year in the U.S., U.K., and France. The device features dual Snapdragon processors, up to four hours of battery life, and a charging case extending usage to 20 hours. The glasses lack external components like pucks or tethers, making them self-contained.
NEA’s Tiffany Luck says enterprises are still figuring out their AI ROI
The article discusses the current state of enterprise AI adoption, highlighting concerns around return on investment (ROI) following initial hype. It features Tiffany Luck, a partner at NEA, who talks about the challenges businesses face in realizing value from AI investments. The piece also touches on recent developments such as companies reducing AI spending and Meta discontinuing its internal AI leaderboard.
FTC lawsuit reveals how subscription scam networks evade app store enforcement
A new lawsuit filed by the U.S. Federal Trade Commission (FTC) details how subscription scam networks like Genesis Tech have evaded app store enforcement by using complex structures involving shell companies and offshore operations. The alleged scheme involved multiple subsidiaries based in Cyprus and operating in Ukraine, marketing various apps to U.S. consumers. These apps reportedly generated nearly $250 million in global revenue between early 2023 and mid-2025, with over $700 million processed through PayPal accounts linked to the network. The case underscores challenges faced by Apple and
Cybercriminals allegedly hacked tens of thousands of Fortinet firewalls used by major companies all over the world
Cybercriminals have allegedly hacked tens of thousands of Fortinet firewalls and VPNs used by major companies worldwide. According to cybersecurity firms Hudson Rock and SOCRadar, the attack does not rely on exploiting unknown vulnerabilities but instead exploits weak password practices by companies. Hackers are scanning the internet for exposed Fortinet devices and using pre-known credentials to gain access. Once inside, they monitor traffic, collect additional credentials, and use them to compromise more devices, creating a self-sustaining cycle.
Social media’s next evolution: user-controlled algorithms
Social media platforms such as Threads, Instagram, and TikTok are introducing features that allow users to customize their algorithmic feeds using AI. This marks a shift from a one-size-fits-all approach to a more personalized experience similar to streaming services. Users can now train their own algorithms to influence what content appears in their feeds, offering greater control over their online experience.
World model maker Odyssey nabs $1.45B valuation backed by Amazon and other big names
Odyssey, a world model AI startup founded by former self-driving vehicle executives, has raised $310 million in a Series B funding round at a $1.45 billion valuation. The round was led by Natural Capital, with participation from Amazon, AMD Ventures, GV, and others. Odyssey uses real-world data to create simulations with accurate physics, similar to how Google Earth gathers information. The company offers world models for various applications, including video game development and robotics. With Amazon's support, Odyssey plans to optimize its models for AWS's Trainium chips.
Mastodon looks to newsletters to help revive the open social web
Mastodon, an open-source social networking platform, has introduced email newsletters as part of its latest update (version 4.6). This feature allows content creators to share their posts with subscribers via email, enabling users to receive updates without needing a Mastodon account. The update also includes improvements such as refreshed user profiles and 'Collections,' which function similarly to suggested follow lists found on other platforms. The goal is to help Mastodon grow its user base by making it easier for people to engage with content without signing up for the service.
Two Stanford grads raise $11M to build a noninvasive wearable for hormone tracking
Two Stanford graduates, Jenny Duan and Abhinav Agarwal, have launched a startup called Clair Health, aiming to develop a noninvasive wearable device that tracks hormone-related health metrics such as inflammation, bloating, energy levels, and menstrual cycle phases. The company recently raised $11.6 million in funding from various investors including Khosla Ventures, a16z speedrun, and others. Clair Health utilizes voice-based onboarding and AI to analyze biomarkers and provide personalized health insights.
Fox enters new era with Roku deal
Fox has agreed to a $22 billion deal to acquire Roku, marking a significant shift in the streaming industry and representing a strategic move for the Murdoch media empire. The acquisition would allow Fox to expand its reach through Roku's platform, which serves over 100 million global households, and combine it with its existing AVOD network Tubi to form a major digital advertising entity. The deal also provides Roku with an exit strategy amid increasing competition from tech giants such as Amazon and Google.
Google bets on Gemini to reinvent the smart home speaker
Google has launched the Google Home Speaker, a new smart speaker designed around its Gemini AI. The device allows users to interact with natural language and make multi-step requests, such as adjusting multiple devices or playing music while setting timers. It supports mid-sentence corrections and includes 10 new voices capable of engaging in two-way conversations.
Anti-G7 protest turns violent as demonstrators torch Tesla and smash UN office windows
Anti-G7 protesters in Geneva turned violent during a demonstration against an upcoming G7 summit in France. Protesters set a Tesla vehicle on fire and smashed windows at a United Nations agency. The protest began peacefully but escalated when demonstrators threw bricks at police, leading to the deployment of tear gas. Demonstrators targeted symbols of capitalism and multilateralism, including the Tesla and the UN building. This follows Elon Musk becoming the world's first trillionaire.
Trump's AI export strategy runs into Trump's export controls
The Trump administration's AI export strategy faces challenges due to internal disagreements over export controls on advanced AI models. A flagship U.S. program aimed at boosting AI exports could be undermined by the same administration that created it. Former AI adviser Dean Ball criticized the government's arbitrary removal of top AI models from foreign use, suggesting the strategy is no longer aligned with current decision-making. Recent export controls on Anthropic's Fable 5 model have caused the company to withdraw access to it, with ongoing discussions between administration officials, 1
PayPal Ventures shutters as company restructuring continues
PayPal Ventures, the corporate venture arm of PayPal, is shutting down as the company continues its restructuring efforts. According to multiple sources, the venture arm—which was established in 2016 and made over 80 investments—will wind down operations. A PayPal spokesperson stated the company is 'exploring strategic options' for the venture arm. The closure comes after the departure of former CEO Alex Chriss, who was replaced by Enrique Lores in February. Critics argue that closing the venture arm may hinder PayPal’s ability to stay competitive in the rapidly evolving fintech landscape.
Collecting robot training data is dirty, unglamorous work. Some AI labs are already paying XDOF to do it.
TechCrunch reports on XDOF, a startup aiming to address the lack of high-quality training data for robotics by providing data pipelines, collection tools, and annotation systems. The article notes that major AI labs are investing in robotics, but face challenges in obtaining the necessary physical interaction data. XDOF has secured $70 million in funding and works with multiple frontier AI labs, though they remain unnamed.
Probably raises $9M to build a more reliable kind of AI
Probably, a startup founded by Peter Elias, has raised $9 million in seed funding from Andreessen Horowitz to develop a more accurate approach to large language models (LLMs). The company aims to reduce hallucinations and factual errors in AI outputs by using a 'mech suit' system that checks LLM responses against a deterministic validator. Its first product is a data science tool that provides answers with citations and audit trails.
Theker just raised $85M to build the factory robot that doesn’t specialize in anything
Theker, an AI robotics startup based in Barcelona, has raised $85 million to develop adaptable factory robots capable of performing a variety of tasks rather than being limited to specialized functions. The company aims to address labor shortages in manufacturing by creating flexible robotic systems that can be reconfigured for different tasks such as sorting packages, packing clothing, or handling bottles and cans. Theker's approach contrasts with traditional humanoid robots designed for specific roles, offering a more versatile solution for complex industrial environments.
Canadian pension giant joins race to fund India’s AI-fueled data center boom
Canadian pension fund CPP Investments has committed up to ₹70 billion ($741 million) to Indian data center operator CtrlS, acquiring an 8.2% stake and forming a joint venture to develop hyperscale data center campuses in India. The investment highlights growing global interest in India's role in supporting AI and cloud infrastructure. CtrlS, based in Hyderabad, currently operates over 15 data centers and is expanding to meet rising demand from cloud providers and AI workloads.
DeepL acquires Mixhalo for live-event audio streaming and translation
DeepL has acquired Mixhalo, a real-time audio startup focused on improving live event experiences through audio streaming and translation. Mixhalo was originally founded in 2016 with the goal of enhancing the listening experience at concerts but later expanded into real-time audio solutions for sports and other live events. The startup had raised over $39 million from various investors. According to Mixhalo's CEO, the growth of voice AI technology has created competitive pressures that made the acquisition more strategic.
Pinterest launches an experimental AI shopping app called ‘Ask Pinterest’
Pinterest has launched an experimental AI shopping app called 'Ask Pinterest,' which allows users to interact with the platform through a conversational, chatbot-like interface. The app aims to enhance product discovery by leveraging Pinterest's 'Taste Graph' data. The announcement was made ahead of the Cannes Lions advertising conference, which focuses on AI's role in marketing. Pinterest also introduced other AI-related initiatives for advertisers.
Uber will bring its premium robotaxi service to Houston in 2027
Uber plans to launch a premium robotaxi service in Houston by mid-2027 as part of its partnership with EV manufacturer Lucid and autonomous vehicle startup Nuro. This follows similar efforts in the San Francisco Bay Area, where the three companies aim to roll out the service later this year. Uber intends to expand the robotaxi program to 'dozens of cities' in the future. In both San Francisco and Houston, Uber will compete with Waymo, an Alphabet-owned autonomous vehicle company already operating robotaxi services in those areas. Testing is ongoing in Houston with a fleet of 100 autonomous车辆,仍
Apple plans to change its Hide My Email privacy feature that could make it less effective
Apple is changing its Hide My Email feature by moving anonymized email addresses from the @icloud.com domain to @private.icloud.com. This change could make it easier for apps and websites to identify and potentially block anonymous sign-ups. Existing addresses will still function, but app and email providers must update their systems to accommodate the new domain. Some users have expressed concern over this change.
OpenAI files confidential SEC paperwork for IPO, opening the door to a Wall Street debut
OpenAI has filed preliminary, confidential paperwork with the U.S. Securities and Exchange Commission (SEC), signaling its potential move toward an initial public offering (IPO). This development positions OpenAI as part of a growing group of major artificial intelligence companies preparing for Wall Street debuts, including rivals like Anthropic and SpaceX. OpenAI CEO Sam Altman previously indicated that an IPO could be the most viable path for the company due to its size and the substantial capital required for technological advancement. The company emphasized that no final decisions haveyet
SpaceX passes Amazon as valuation balloons to $2.7T
SpaceX surpassed Amazon to become the fifth-most valuable company in the world, with its valuation reaching $2.7 trillion following a significant rise in its stock price. This increase occurred despite SpaceX reporting a $4.9 billion loss on $18.7 billion in revenue in 2025, while Amazon reported a $78 billion profit on $717 billion in sales. SpaceX has expanded its revenue through compute leasing agreements with Anthropic and Google and recently acquired AI coding startup Cursor in a $60 billion all-stock deal. The company's initial public offering (IPO) valued it at approximately $1.7万亿, and
Qualcomm wants to be the chip inside whatever replaces your smartphone, and it just announced two products toward that end
Qualcomm is expanding beyond traditional mobile chips by developing AI-powered wearable devices such as smart glasses, earbuds with cameras, and other accessories. The company has introduced two key products: the Snapdragon Reality Elite, an advanced platform for mixed-reality glasses with enhanced AI capabilities, and the START toolkit, which combines hardware and software for AI-ready devices. These innovations aim to position Qualcomm as a leader in the next generation of computing platforms beyond smartphones.
Bug in FIFA World Cup internal system gave anyone ability to modify TV stream
A security researcher known as BobDaHacker discovered a vulnerability in FIFA's internal systems that allowed unauthorized access to the TV streams of World Cup games. By registering as a player agent on FIFA's official platform, the researcher exploited a flaw in the back-end API that failed to verify user authorization. This enabled access to internal systems used by broadcasters to control what appears on viewers' TVs and commentators' screens. The researcher highlighted the potential risk of an attacker hijacking all cameras or disrupting the broadcast. FIFA addressed the issue shortly之后,但
Mobileye’s US robotaxi launch will put it on both sides of the AV business
Mobileye, an Intel subsidiary and provider of autonomous vehicle technology, announced plans to launch a robotaxi service in a U.S. city by 2027. This marks a shift from its role as a supplier of autonomous driving technology to becoming an operator in the robotaxi space. The company did not specify the location but mentioned an initial fleet of 100 autonomous vehicles, with plans to expand to approximately 17,000 robotaxis within five years. Mobileye's CEO emphasized the growing importance of technology providers in the autonomous vehicle industry.
Rivian cuts hundreds of workers after R2 deliveries start
Rivian has announced layoffs affecting less than 2% of its workforce, citing the need to improve efficiency as it works toward profitability. This marks the fourth round of layoffs since early 2024. The company delayed its target for turning a profit due to high costs associated with developing autonomous vehicle technology. Rivian also disclosed that Uber plans to invest up to $1.25 billion and purchase up to 50,000 R2 SUVs for use as robotaxis.
Payments startup Flutterwave hits $3.2B valuation, backed by Ripple
Flutterwave, an African payments infrastructure company, has reached a $3.2 billion valuation following a Series E funding round that includes an equity investment from Ripple. The funding aims to support expansion of financial services across Africa, where cross-border payments face challenges such as fragmented banking systems and currency volatility. Flutterwave has previously acquired the African banking startup Mono and partnered with Polygon Labs to introduce stablecoin solutions for faster and cheaper transactions.
Plaud says its software business topped $100M in ARR after shipping over 2M AI notetakers
Plaud, a company that produces AI-powered notetakers, announced that its software business has achieved over $100 million in annual recurring revenue (ARR). The company claims to have sold more than 2 million of its devices, including models like the Plaud Pins and credit-card-sized gadgets designed to attach to phones. Plaud emphasizes that its devices enable users to engage in real-life conversations rather than relying solely on digital inputs. The company highlights its unique approach compared to other AI firms that focus on software behind screens. Recently, Plaud launched new products,
Robinhood’s note on 10% layoffs shows blaming AI isn’t cutting it
Robinhood announced a 10% reduction in its workforce, approximately 290 full-time employees, without attributing the layoffs to AI. The company described the move as part of a restructuring effort rather than a response to AI advancements. CEO Vlad Tenev emphasized the importance of maintaining a 'lean, hyper-focused' team structure, aligning with broader trends among other tech firms such as Amazon, Block, Coinbase, GitLab, and Intuit.
Threads adds new personalization and community features as it reaches 500M monthly users
Threads, a social media platform owned by Meta, has reached 500 million monthly active users and introduced new features such as 'Your Algo' and 'Communities.' 'Your Algo' allows users to privately customize their feed content for a set period, while 'Communities' provides dedicated spaces for focused discussions.
India orders temporary ban on Telegram over exam fraud concerns
India's National Testing Agency has ordered a temporary ban on Telegram until June 22 due to concerns that the messaging app is being used to facilitate exam fraud ahead of the re-test of the National Eligibility Entrance Test (NEET (UG)). The agency claims that Telegram is being used to sell fake exam papers and spread misinformation. Additionally, Telegram is required to disable its message-editing feature until June 30. The order was issued under Section 69A of India's Information Technology Act. The decision has faced criticism from digital rights advocates.
ChatGPT’s market share slips below 50% for first time
ChatGPT's global market share has dropped below 50% for the first time, according to Sensor Tower's State of AI Report for 2026. Competitors such as Google's Gemini and Anthropic's Claude have gained significant ground, with Gemini holding 27.7% and Claude 10.3%. OpenAI reports 900 million weekly active users, while ChatGPT maintains the highest number of monthly users at 1.1 billion.
Sundar Pichai faces boos, walkout at Stanford graduation ceremony over Google’s Israel, ICE ties
At Stanford University's graduation ceremony, Google CEO Sundar Pichai faced student protests during his commencement speech. Approximately 200 students walked out, and others booed him, criticizing Google's involvement in defense contracts such as Project Nimbus, which provides cloud and AI services to the Israeli military, and its relationship with U.S. Immigration and Customs Enforcement (ICE). Protesters held signs with messages like 'ICE SPIES WITH GOOGLE AI' and 'GENOCIDE RUNS ON GOOGLE,' and waved Palestinian flags. The walkout was organized by campus activist groups, including Stanford
Meta’s new ‘AI Mode’ on Facebook pulls from public info across its platforms
Meta has introduced 'AI Mode' on Facebook, allowing users to ask questions in plain language and receive synthesized answers derived from public posts across the platform, including Groups and Reels. This comes after the quiet launch of Forum, a Reddit-style app with an AI-powered 'Ask' tab. The features aim to enhance user interaction and content creation but raise concerns about the reliability of information sourced from unvetted user-generated content.
Salesforce acquires AI customer service platform Fin for $3.6B
Salesforce has announced the acquisition of Fin, an AI customer service platform formerly known as Intercom, for $3.6 billion. Fin provides AI agents capable of resolving customer inquiries through multiple communication channels. Salesforce plans to integrate Fin's technology and team into its existing enterprise platform, Agentforce, to enhance its AI agent capabilities. The acquisition is expected to finalize in early 2027. Fin's co-founder and CEO expressed excitement about the partnership and the potential for growth.
Sarvam becomes India’s newest AI unicorn with $234 million funding round led by HCLTech
Sarvam, an AI startup based in Bengaluru, India, has secured $234 million in funding, achieving a $1.5 billion valuation and becoming India's newest AI unicorn. The funding includes $150 million from HCLTech, with additional contributions from Bessemer Venture Partners, Khosla Ventures, and Peak XV Partners. This follows previous investments totaling $41 million. Sarvam plans to raise up to $300 million in its Series B round. The company focuses on developing AI models tailored for Indian languages and deploying them across various industries such as banking, insurance, and government services
As AI agents become employees, NewCore emerges with $66M to give them identities
Cybersecurity startup NewCore has raised $66 million in funding to address challenges related to authenticating, governing, and controlling AI agents as they are increasingly treated as workplace participants. The company aims to provide identity solutions for these digital workers, similar to how human employees are managed. NewCore’s approach is driven by the belief that current identity systems are inadequate for managing the growing presence of AI in enterprises.
A satellite just learned to find things on its own — here’s what that means
A satellite, YAM-9, successfully used a vision-language model (VLM) developed by Google DeepMind called Gemma 3 to identify areas of interest based on natural language queries, marking the first use of such technology in orbit. This advancement allows satellites to process data independently, reducing reliance on ground-based analysts and improving efficiency. The VLM combines image analysis with language understanding, enabling tasks like identifying infrastructure near railway hubs or classifying environmental changes.
Startup CEO Charlie Javice is reportedly angling for a Trump pardon
Charlie Javice, the former CEO of Frank and a convicted fraudster, is reportedly seeking a presidential pardon. According to the Wall Street Journal, she is reaching out to individuals connected to the Trump administration. The White House is considering granting around 250 pardons ahead of America's 250th anniversary. Javice was found guilty of creating fake customer accounts to inflate her startup's valuation before selling it to JPMorgan Chase for $175 million. She is currently serving a prison sentence and appealing her conviction. JPMorgan had previously closed accounts linked to Donald J
Apple’s Best New App in Years Does Just One Job
The article critiques recent trends in app development where technology companies add features that users did not explicitly request, often leading to cluttered interfaces. The author highlights examples such as AI chatbots in email clients, AI-driven search bars, and additional tools like 'Image Playground' on new iPhones. The piece also discusses sports apps, noting that while they provide essential information like scores and videos, they increasingly include unnecessary features that detract from their core purpose. The article praises Apple for creating a simple, focused app that does one
The new Sonos Play has become my go-to desk and kitchen speaker
The author reviews the new Sonos Play, a hybrid desktop and portable speaker. They highlight its compact design, portability, and ease of use for tasks like listening to podcasts while moving between rooms. Features such as built-in voice assistants, physical controls, and the ability to stay aware of surroundings are noted as advantages. However, the author points out some minor usability issues, including the difficulty in locating the controls due to their subtle design.
Meta reportedly moves to unwind $2B Manus deal after Beijing’s demand
Meta begins to dismantle its $2 billion acquisition of Manus following a directive from Beijing to reverse the deal.
OpenAI faces investigation from state attorneys general
OpenAI is facing an investigation by multiple state attorneys general. The scope of the inquiry includes topics such as OpenAI's advertising policies and how it handles health data.
This thin under-pillow speaker helped me fall asleep without earbuds
The article discusses the author's experience with insomnia and their preference for using audio content like rain sounds to aid sleep. The author mentions discomfort with earbuds and the inconvenience of playing audio out loud. They introduce a thin under-pillow speaker as a solution.
The FBI built its own replica small town to simulate real-world cyberattacks
The FBI has created a 22,000 square-foot replica town on its Huntsville, Alabama campus called the Kinetic Cyber Range. This facility is used to train law enforcement in simulating and investigating real-world cyberattacks. The replica includes fully furnished houses, a hotel, a gas station, a grocery store, a courthouse, a hospital, and a power company, all wired with functioning devices and systems that mimic real-world environments. The FBI reports that the facility has trained over 1,400 students, including FBI personnel and partners from other federal and local agencies. The training aims to provide hands-on experience with consumer and enterprise technologies that are frequently targeted by hackers. The FBI's 2025 Internet Crime Report highlights a significant increase in cybercrime losses, with ransomware being the top ongoing threat to critical infrastructure.
Andrew Yang thinks the next big startup opportunity is lowering the cost of living
Andrew Yang, a former presidential candidate and entrepreneur, suggests that the next major startup opportunities lie in reducing the cost of living. He highlights examples such as Cost Plus Drugs and his own venture, Nobile Mobile, which offers affordable cell service and refunds unused data. Yang argues that as AI threatens to reduce wages and displace workers, there will be growing demand for services that lower essential living costs.
Meta’s months-old AI unit is a soul-crushing gulag, say the engineers stuck inside it
A new report indicates that Meta's AI unit, employing 6,500 people, is facing potential internal unrest among its engineers.
Chinese cybercrime operation that used AI to scam ‘hundreds of thousands of victims’ sued by Google
Google has filed a lawsuit against a Chinese cybercrime group known as 'Outsider Enterprise' for using AI to scam hundreds of thousands of victims through the distribution of 2.5 million text messages over a two-week period.
Mistral is rumored to be raising €3B at €20B valuation
Mistral AI is reportedly raising €3 billion in a funding round that would value the company at approximately €20 billion, nearly doubling its previous Series C valuation of €11.7 billion.
Robinhood sees ‘record-breaking’ traffic after SpaceX stock debuts
Robinhood saw 'record-breaking' traffic following the debut of SpaceX stock, though some users faced temporary service disruptions which have since been resolved.
Opendoor’s India exit is fueling a bigger conversation about AI and outsourcing
Opendoor, a San Francisco-based online home-buying platform, is closing its India operations less than two years after entering the market. The move has sparked discussions about how AI might be changing the economic dynamics of offshore work. Opendoor's CEO mentioned a strategy to bring operations closer to its customer base in the U.S. and a focus on smaller AI-native teams. The closure has drawn attention in Silicon Valley, where it is seen as an early indicator of AI's impact on the global outsourcing industry. India, which has grown into a major center for multinational offshoring, hosts
Apple introduces systemwide dictation
Apple introduced a new systemwide dictation feature during its Worldwide Developers Conference (WWDC 2026), integrated into the keyboard and capable of correcting spelling, punctuation, and capitalization across apps. The feature is part of Apple's new Apple Intelligence model, which is based on Google's Gemini AI. The article notes that Apple's in-built dictation could offer improved ease of use compared to existing third-party apps such as Wispr Flow, Willow, and Monologue, which also provide transcription services but require additional setup steps. Apple has not yet confirmed whether it'll
US surveillance law to expire for first time after lawmakers reject Trump’s controversial pick to lead spy agencies
The US surveillance law known as Section 702, which authorizes the NSA and FBI's warrantless surveillance, will all but certainly expire on Friday for the first time.
Cheaper, faster, and culturally aware, Avataar’s video AI is built for India’s scale
Avataar AI, a startup participating in India's $1.2 billion India AI Mission, has developed a video AI model called Varya. Built upon a publicly available model from Alibaba, Varya uses distillation to become significantly faster and cheaper while being tailored to Indian cultural contexts such as festivals, food, and clothing. The model generates videos much quicker and at lower costs compared to the original model.