Latest AI News
Clawdmeter turns your Claude Code usage stats into a tiny desktop dashboard
Silicon Valley’stokenmaxxingera now has its own hardware. A new open source project brings your Claude Code utilization stats into a tiny desktop dashboard, allowing AI power users to keep an eye on their usage. Sure, you can track Claude Code usage directly in the terminal using commands or otherexternal toolsandapps, but that’s not as fun as seeing a pixel-art version of the Clawd sprite dance on a screen before showing at-a-glance token usage info, is it? The “Clawdmeter,” as the device is called, is both a fun side project for AI power users and a timely indication of how thoroughly Anthropic’s Claude has infiltrated the developer community and the growing interest in tokenmaxxing. This new “productivity” trend sees software engineers at various tech companies maximizing the number of AI tokens consumed at work as a measure of how much they’ve embraced AI. As one Reddit userjokedwhen seeing theprojectfor the first time: “At this point, Anthropic should just mail these to us for free.” Another suggested adding a button to increase capacity or top up more tokens using your card on file. (Ha, that could be dangerous!) Someone made a small hardware device for the desk that monitors Claude token usage.Calls it Clawdmeter. Kinda genius.pic.twitter.com/COD4EmmsTy The idea for the project comes from Reykjavik, Iceland-based software developerHermann Haraldsson, who says he had always wanted to play around with embedded devices, but never before had the time. “I’m not an embedded developer or anything like that,” Haraldsson told TechCrunch in a call. But Claude was able to walk him through the project in just a few days, he said. “It’s really democratized access to programming, so that anyone can now do what developers used to do. I think that’s really positive, actually.” Most of the time he spent building the device was focused on design, making sure to get the font, colors, and little animations just right. To build your own dashboard, you can use a small lithium-ion battery-powereddisplaylike theWaveshare ESP32-S3-Touch-AMOLED-2.16, which pairs with your laptop over Bluetooth. When the device is turned on, the splash screen playspixel-art Clawd animationsthat get busier as your usage rate climbs. You can also push the middle button to cycle through different types of animations if you choose. “I like it when I’m working, and I see it going crazy — it’s like a little dopamine loop,” notes Haraldsson. The animation remains on the screen until you press the middle button, which then displays your session and weekly Claude utilization data in simple charts. You can press this button again to cycle to the Bluetooth screen, which displays the connection status and offers a reset feature. From there, you can tap the screen to return to the original splash screen animation. Meanwhile, two other side buttons send Space and Shift+Tab over Bluetooth for Claude Code’s voice mode and mode-toggle shortcuts. The latter lets you move between the default Normal mode, “Accept Edits” mode, Plan Mode, and Auto Mode. Haraldsson says the device keeps up with your usage limits because it reads your Claude Code OAuth token to make an API call, which then pulls the usage numbers from the response headers directly. Because Clawdmeter is an open source project, anyone can fork it to add their own features, animations,screens, and more, based on their particular interests and needs. Haraldsson says he was surprised to see that over 800 people have starred it on GitHub since its May 10 launch, and 50 have already forked the project for their own development. He suspects the device appeals to them because it has a nostalgic feel. “There’s a kind of nostalgia for when you used to have a hardware device for everything — like a Walkman to play music, or an iPod,” says Haraldsson. (Or, as one Redditorput it, the Clawdmeter is like a “hardware Tamagotchi for my context window.”) “I know it’s not replacing anything — like, you could have this on your computer — but it’s just fun,” Haraldsson says.
View
OpenAI is reportedly preparing legal action against Apple; it wouldn’t be the first partner to feel burned
OpenAI is so frustrated with Apple over a ChatGPT integration that failed to deliver the subscribers and prominence it expected that the company is now actively exploring legal action against the iPhone maker,Bloomberg News reported Thursday, citing people familiar with the matter. According to Bloomberg, OpenAI has enlisted an outside law firm to work through its options, which could include sending Apple a formal breach-of-contract notice without necessarily escalating to a full lawsuit (at least not immediately). Any legal move would likely wait until after the conclusion of OpenAI’s ongoing trial with Elon Musk. Still, it’s a reminder of what a difficult partner Apple can be for major software companies. The iPhone is an enormously attractive platform for growth, but it’s fully under Apple’s control — and companies that build there are only guests. From Google to Adobe, there’s a long history of Apple showing guests the door when they seem as if they’re getting too comfortable. TechCrunch has reached out to both OpenAI and Apple for comment. The OpenAI partnership, announced at Apple’s Worldwide Developers Conference in June 2024, wove ChatGPT into Apple’s operating systems as an option within Siri and as part of the iPhone’s Visual Intelligence feature (allowing users to use their camera to analyze their surrounds and send photos to ChatGPT with related questions). OpenAI, along with industry watchers, expected the deal might eventually funnel billions of dollars in new subscriptions its way and give the company prime real estate across one of the world’s most-used mobile ecosystems. Instead, Bloomberg reports, OpenAI has grown increasingly aggravated, complaining that the integration has been buried, its features hard to find, and that revenue from the tie-up is nowhere close to projections. “They basically said, ‘OpenAI needs to take a leap of faith and trust us,’” one OpenAI executive told Bloomberg. “It didn’t work out well.” Apple, for its part, has its own grievances, including concerns about OpenAI’s privacy standards and, according to Bloomberg, irritation over OpenAI’s push into hardware, an effort led by former Apple executives including ex-design chief Jony Ive. Either way, OpenAI is hardly the first partner of Apple to regret hitching its wagon to the company. Apple has a long history of embracing partners and then alienating them. The most famous case is Google Maps, which was a flagship feature of the original iPhone. It was so central to the device's appeal that its removal in 2012 — replaced by Apple's markedly inferior Apple Maps product — became one of the biggest tech fiascos of the decade, prompting a rare public apology from CEO Tim Cook. The friction between the two companies had been building for years at that point, thanks to the rollout of Google's Android phone a year after the iPhone's 2007 debut; after Google's then-CEO Eric Schmidt stepped down from Apple's board in 2009, that rivalry only intensified. Adobe has some scar tissue, too. Steve Jobsrefused to support Flashon the iPhone and iPad, publishing a famousopen letterin 2010 explaining why and effectively dooming the technology. Flash never recovered its footing on mobile. Then there's Spotify, whichspent yearsarguing that Apple leveraged its control over the App Store to disadvantage rival music streaming services after launching Apple Music in 2015. The European Commission agreed, fining Apple nearly€1.8 billionin March 2024. Sometimes these rifts can be overcome in the name of commercial interests. Google is now Apple's AI infrastructure partner, having struck amultiyear dealin January to power the next generation of Apple Intelligence with Gemini models. Apple is paying Google roughly $1 billion a year. In the meantime, OpenAI has had its own share of strained relationships lately. Elon Musk'slawsuitagainst the company — which accuses OpenAI ofabandoningits nonprofit founding mission and operating inbad faith— is currently at trial. The company has also reportedly navigated tensions with Microsoft, its biggest backer and infrastructure partner, as it pushes for greater independence ahead of its own IPO ambitions.
View
What happens when AI starts building itself?
Richard Socher has been a major figure in AI for some time, best known for founding the early chatbot startup You.com and, before that, his work on Imagenet. Now, he’s joining the current generation of research-focused AI startups with Recursive Superintelligence, a San Francisco-based startup that came out of stealth on Wednesday with $650 million in funding. Socher is joined in the new venture by a cohort of prominent AI researchers, including Peter Norvig and Cresta co-founder Tim Shi. Together, they’re working to create a recursively self-improving AI model, one that can autonomously identify its own weaknesses and redesign itself to fix them, without human involvement — a long-held holy grail of contemporary AI research. I spoke with him on Zoom after the launch, digging into Recursive’s unique technical approach and why he doesn’t think of this new project as a neolab, the informal term for a new generation of AI startups that prioritize research over building products. This interview has been edited for length and clarity. We hear a lot about recursion these days! It feels like a very common goal across different labs. What do you see as your unique approach? Our unique approach is to use open-endedness to get to recursive self-improvement, which no one has yet achieved. It’s an elusive goal for a lot of people. A lot of people already assume it happens when you just do auto-research. You know, you can take AI and ask it to make some other thing better, which could be a machine learning system, or just a letter that you write, or, you know, whatever it might be, right? But that’s not recursive self-improvement. That’s just improvement. Our main focus, is to build truly recursive, self-improving superintelligence at scale, which means that the entire process of ideation, implementation and validation of research ideas would be automatic. First [it would automate] AI research ideas, eventually any kind of research ideas, even eventually in the physical domains. But it's particularly powerful when it's AI working on itself, and it's developing a new kind of sense of self awareness of its own shortcomings. You used the term open-ended — does that have a specific technical meaning? It does. In fact, Tim Rocktäschel, one of our cofounders, led the open-endedness and self-improvement teams at Google DeepMind and particularly worked on the world model Genie 3, which is a great example of open-endedness. You can tell it any concept, any world, any agent, and it just creates it, and it's interactive. In biological evolution, animals adapt to the environment, and then others counter-adapt to those adaptations. It's just a process that can evolve for billions of years, and interesting stuff keeps happening, right? That's how we developed eyes in our [heads]. Another example is rainbow teaming, fromanother paper from Tim. Have you heard of red teaming? In cybersecurity, it means-- So, red teaming also has to be done in an LLM context. Basically you try to get the LLM to tell you how to build a bomb, and you want to make sure that it doesn’t do it. Now, humans can sit there for a long time and come up with interesting examples of what the AI shouldn't say. But what if you tested this first AI with a second AI, and that second AI now has the task of making the first AI [try to] say all the possible bad things. And then they can go back and forth for millions of iterations. You can actually allow two AIs to co-evolve. One keeps attacking the other, and then comes up with not just one angle but many different angles, and hence the rainbow analogy. And then you can inoculate the first AI, and you become safer and safer. This was an idea from Tim Rocktaeschel, and it’s now used in all the major labs. How do you know when it’s done? I suppose it’s never done. Some of these things will never be done. You can always get more intelligent. You can always get better at programming and math and so on. There are some bounds on intelligence; I’m actually trying to formalize those right now, but they’re astronomical. We’re very far away from those limits. As a neolab, it feels like you’re supposed to be doing something that the major labs aren’t doing. So part of the implication here is that you don’t think the major labs are going to reach RSI [recursive self-improvement] by doing what they’re doing. Is that fair to say? I can’t really comment on what they’re doing, but I do think we’re approaching it differently. We really embrace the concept of open-endedness, and our team is entirely focused on that vision. And the team has been researching this and doing papers in this space for the last decade. And the team has a track record of really pushing the field forward significantly and shipping real products. You know, Tim Shi built Cresta into a unicorn. Josh Tobin was one of the first people at OpenAI and eventually led their Codex teams and the deep research teams. I actually sometimes struggle a little bit with this neolab category. I feel like we're not just a lab. I want us to be become a really viable company, to really have amazing products that people love to use, that have positive impact on humanity. So when do you plan to ship your first product? I’ve thought about that a lot. The team has made so much progress, we may actually pull up the timelines from what we had initially assumed. But yes, there will be products, and you’ll have to wait quarters, not years. One of the ideas around recursive self-improvement is that, once we have this sort of system, compute becomes the only important resource. The faster you run the system, the faster it will improve, and there’s no outside human activity that will really make a difference. So the race just becomes, how much processing power can we throw at this? Do you think that’s the world we’re headed toward? Compute is not to be underestimated. I think in the future, a really important question will be: how much compute does humanity want to spend to solve which problems? Here’s this cancer and here’s that virus — which one do you want to solve first? How much compute do you want to give it? It becomes a matter of resource allocation eventually. It’s going to be one of the biggest questions in the world.
View
OpenAI says Codex is coming to your phone
Codex is going mobile. The coding tool — which OpenAI launchedapproximately a year ago— has now been integrated into the ChatGPT app, allowing users to monitor and manage their development workflows remotely. The new function allows users to see their Codex live environments in any devices where it is running. The company announced the changes Thursday; the update, which is currently in preview, is now available to all plans on iOS and Android. “This is more than the ability to remotely control a single task or dispatch new tasks to your computer,” OpenAI said in a statement. “From your phone, you can work across all of your threads, review outputs, approve commands, change models, or start something new.” Last month, OpenAI alsogave Codex the abilityto run in the background in desktop environments — empowering the tool to take care of various tasks autonomously. Earlier this month, the companyalso introduceda Chrome extension that allows the agent to work in live browser sessions. In February, Anthropic released a similar feature —Remote Control— which allows users to remotely monitor Claude Code’s work from afar. The flurry of feature releases from both OpenAI and Anthropic speaks to the tense competition between the two over whose agentic coding tool will become the most widely used. Over the past year, Anthropic’s Claude Code hasgained in popularityamong businesses and tech professionals alike, although both tools continue to be widely used.
View
Wirestock raises $23M to supply creative multi-modal data to AI labs
In the past few years, creative marketplaces and platforms have realized that they are sitting on a data goldmine, and they can either use that data to develop AI models themselves or turn it into a source of revenue by licensing it to other AI labs. Wirestock, which previously helped photographers distribute and sell their work on stock photography services like Shutterstock, has taken the latter path. The company pivoted to being a data provider in 2023, and now supplies datasets of images, videos, design assets, and gaming and 3D content to AI labs. Wirestock said its platform has signed up more than 700,000 artists and designers who complete different tasks for data collection, similar to freelancers on platforms like Fiverr. Mikayel Khachatryan, the company’s co-founder and CEO, said Wirestock was transparent about its shift, and allowed artists to opt out of its data supply business (in 2022, the platform had over100,000 photographerson its platform). Khachatryan didn’t specify how many people switched over as data providers for AI, only saying “the majority” did. “Initially, a lot of our deals were just selling what we had off the shelf, like our existing library. But then it turned into a lot of custom requests for content and data, and that created new opportunities for creators, and the platform just took off,” he said. The startup said on Thursday it has raised $23 million in Series A funding to build out the new data supply business. The round was led by Nava Ventures, and saw participation from SBVP (co-founded by Sheryl Sandberg), Formula VC, and I2BF Ventures. Khachatryan says Wirestock currently provides multi-modal data to six of the largest foundation model makers, but he wouldn’t name them. He noted the company currently has an annual run-rate revenue of $40 million and has so far paid out $15 million to its contributors. As part of the transition, the startup had to retrain some of its teams to annotate and label data in detail to make it useful for AI labs. It also had to build sales and enterprise teams to be able to pitch to hyperscalers, and find ways to get more creative assets in areas like 3D modeling. Wirestock currently uses email marketing and referral programs to bring in new contributors. Photographers, videographers, and illustrators can apply to provide data on its website, but they have to complete an unpaid task as a quality check before they're accepted. The company said it uses a mix of AI and human reviews to evaluate all work on the platform. Demand for data supply services is sky high right now as AI labs race to continue improving their models. Companies like Surge, Scale AI, and Mercor have built businesses worth tens of billions seemingly overnight on the back of demand for different kinds of datasets, and a slate of new startups such asMicro1,Human Archive, andHuman Native AIare working with top AI model makers, too. Wirestock wants to focus on providing data for models that aid creative use cases, such as image and video generation. The company is also exploring other modalities like audio and music. Freddie Martignetti, founder of Nava Ventures, said his fund was looking for a startup that was innovating on data procurement and refinement even before it learned about Wirestock. "I think Wirestock has a deep understanding of what foundational models and hyperscalers need in terms of multi-modal data to start creating more human-like systems. The cornerstone of our thesis was that multi-modal data will be increasingly important, not just to create images or videos, but for models to complete real-world tasks," Martignetti told TechCrunch. Wirestock currently employs 60 people, and will use the new funding to hire for research, engineering, and product roles. It is also building enterprise software for AI labs to collaborate on datasets. The funding round brings the startup's total capital raised to about$26 million.
View
Two weeks left: Startup Battlefield 200 applications close May 27
Your shot at VC access, global visibility, TechCrunch coverage, and $100,000 equity-free funding is running out. Startup Battlefield 200applications officially close in just two weeks. If you’re building a breakout startup — or know a founder who is — now is the time to act. Apply by May 27for the opportunity to take the stage atTechCrunch Disrupt 2026alongside 200 of the world’s most promising early-stage startups. Pre-Series A founders, consider this your countdown reminder: The strongest startups are already entering the arena. If your startup has already been nominated, don’t wait to complete your application. The final stretch moves fast, and late submissions risk getting lost in the noise. Know a startup that deserves the spotlight?Nominate them nowso they still have time to apply before the deadline. Startup Battlefield 200 is where breakout companies get discovered. Selected startups will showcase live at Disrupt 2026 in front of 10,000+ attendees, leading VCs, global media, and the full TechCrunch audience. This is your chance to gain investor exposure, receive direct VC feedback, and prove your company belongs among the next generation of category-defining startups. Early applicants gain an advantage: more time to prepare, more opportunities to stand out with TechCrunch editors, and more momentum before competition intensifies. We’re looking for ambitious early-stage startups building innovative, potentially category-defining products. Applications are open globallyacross all industries. Most selected companies are pre-Series A, though select Series A startups may qualify on a case-by-case basis. To apply, startups should have: This is the same launchpad that helped accelerate companies like Dropbox, Discord, Fitbit, Trello, and Mint. Thousands apply every year. Only 200 are selected. Just 20 finalists pitch live on the Disrupt Stage. One startup takes the crown. Startup Battlefield 200 offers one of the highest-ROI opportunities available to early-stage founders — andit’s free to apply. Selected startups receive: Two weeks. That’s it. The founders who break through aren’t waiting for the final hour — they’re already making their move. If you’re building something category-defining, or know a startup that deserves the spotlight,submit your nomination and complete your application before May 27.
View
Cisco cuts nearly 4,000 jobs to spend more on AI, reports ‘record quarterly revenue’
Technology giant Cisco is cutting fewer than 4,000 jobs, or around 5% of its workforce, despite reportingbetter-than-expected profit and revenuein its fiscal third quarter. The networking equipment maker said it reducing its headcount in order to change its “cost structure” and invest in AI and cybersecurity. Cisco’s decision follows a recent trend of tech companies increasingly citing a priority on AI spending as a reason to let employees go.CloudflareandGeneral Motorshave both laid off staff in recent days, despite reporting strong financial results. Cisco said it plans to invest more in cybersecurity, as the company continues to contend witha slewofsecurity vulnerabilitiesin its routers and firewalls that have allowed hackers to break into the networks of its corporate customers,including the U.S. government. Cisco last year also experienceda data breachin which customers’ personal information was affected. Ina blog postpublished Wednesday, Cisco’s chief executive Chuck Robbins touted the company’s “record revenue” and “double-digit growth,” while acknowledging that the company was making strategic investments “in our employees’ use of AI across the company.” Accordingto public filings, Robbins was slated to earn more than $52 million in executive compensation during 2025. When reached by TechCrunch, a Cisco spokesperson did not comment beyond Robbins’ statement, or say, when asked, if Robbins plans to reduce his compensation. This is the latest round of job cuts at Cisco in recent years. The company laid off thousands of employeesduring two separate layoffs in 2024, andcut 150 jobsin 2025.
View
Khosla Ventures is betting $10M on Ian Crosby, whose first startup, Bench, imploded
Ian Crosby, whose previous startup Bench Accounting famouslyshut down in 2024before beingbought for scraps, is taking another shot at building a business out of automating the arduous work of bookkeeping. His new startup,Synthetic, aims to build a fully autonomous AI bookkeeper that can generate accrual-based financials without direct human involvement. Although the product is still in the design phase — and Crosby admits his vision may not yet be technologically possible — the startup has raised $10 million in a seed funding round led by Khosla Ventures, with participation from Basis Set Ventures and Shopify CEO Tobias Lütke. Most investors would run from a founder facing the kind of challenges Crosby is right now — the fallout of his previous business collapsing and a vision that may exceed the technical feasibility of current foundational models. But Khosla partner Jon Chu told TechCrunch he sometimes does just the opposite: “I tend to run towards controversy a little bit.” “In controversy, groupthink often shapes the narrative rather than the truth of the story itself,” he said, citing Parker Conrad’s 2016 ousting from Zenefits as an example. While the industry narrative was initially critical of Conrad, he subsequently founded Rippling, which is now valued at nearly $17 billion. “I believe people have room for growth,” Chu said about his bet on Crosby and Synthetic. Crosby maintains he wasn’t directly responsible for bringing Bench to the point of insolvency. According to him, he wasfired by Bench’s boardin 2021, three months after he turned down a $250 million acquisition offer from Brex. The board also disagreed with Crosby’s strategic direction, especially as the business was bleeding cash, and his executive team was reportedlyfrustrated with his direct leadership style. “He took a big swing, made a few mistakes. That didn’t go well,” Chu said. Bench ultimately imploded when its new management proved unable to restore the company to health on its own. After leaving Bench, Crosby joined Shopify and founded Teal, another accounting startup, which was bought by Mercury 18 months later. As part of his due diligence, Chu said he spoke with several executives who worked with Crosby after his departure from Bench, and they all “had fantastic things to say about Ian,” Chu told TechCrunch. Chu is convinced that the three roles Crosby held after leaving Bench provided the entrepreneur with ample opportunity to learn from his past mistakes. Crosby says he has his eyes firmly set on creating a fully AI-driven bookkeeping service, rather than relying on human accountants, as most accounting startups such as Xero do now. “We're not going to release anything that's not fully autonomous,” Crosby told TechCrunch. “It’s that or bust.” Synthetic plans to serve only AI and other software startups. But Crosby acknowledges that AI models still make significant bookkeeping mistakes. While Synthetic's prototype works for a narrow group of users, he remains uncertain how it will scale for a broader customer base. Crosby explained with an analogy: “It's like a self-driving car that can drive down one street versus the self-driving car that can drive down any street. We haven't driven down enough streets to know if it's going to crash.” Still, the founder says he can afford to be patient and wait for foundational models to become more reliable for bookkeeping calculations. “I've raised years of cash, so we can just wait it out,” Crosby said.
View
Cerebras raises $5.5B, kicking off 2026’s IPO season with a bang
Cerebras raised $5.5 billion in its IPO on Thursday, pricing shares at $185 Wednesday evening, way higher than its range ($115 to $125, later raised to $150-$160), even as it increased the size of the offering to 30 million shares. And pre-market trading indicates that shares are going to open with a giant pop, as retail investors bid up the price to grab them. (We’ll update this story after trading begins.) Even at the IPO price, the company enters its first day of trading at a fully-diluted valuation of $56.4 billion (meaning, accounting for all shares). Co-founder CEO Andrew Feldman’s stake at $185/share is worth nearly $1.9 billion, while co-founder CTO Sean Lie’s stake weighs in at about $1 billion. A year ago, it looked like this day would never happen for Cerebras. The Nvidia competitor, which designed its giant chip from scratch, purpose-built for AI, had first filed to go public in 2024. But concerns about a large investment from Abu Dhabi-based Group 42 mired the IPO in an endless review from the Committee on Foreign Investment in the United States (CFIUS). Investors were also cool about its financials: Group 42 accounted for almost all of Cerebras’s revenues. So those IPO plans were shelved. IPO ambitions reappeared in earnest in April when the companywas able to report about double the revenues: $510 million in 2025 (up 76% year-over-year), and from a handful of customers. It also reported a massive swing to a profit — to $237.8 million in net income — compared to losing nearly half a billion the year before. Investors began salivating. Cerebras has now come out as a major contender for supplying chips for inference — the ongoing compute processing required for models to answer prompts — and now counts OpenAI (in acomplicated circular-deal relationship), G42, Saudi’s Mohamed bin Zayed University of Artificial Intelligence and Amazon Web Services as customers. Developing, will update this post with first-day of trading numbers.
View
Amazon Kills Rufus AI, Replaces It With Alexa for Shopping AI Assistant
Amazon, on Wednesday, announced Alexa for Shopping, a new artificial intelligence (AI) shopping assistant. The new tool comes with agentic capabilities and replaces Rufus AI, a mainstay on the app and the website since February 2024. The Seattle-based tech giant revealed that the new AI experience combines Rufus AI and Alexa+ to bring an assistant that not only has deep knowledge of Amazon's shopping platform and catalogued products, but also brings the immersive conversational layer with agentic automation. The new tool is first rolling out in the US.
View
AI-Native Startup Exaforce Secures $125 Mn to Scale Agentic Security Operations
Beyond geographic expansion, the company intends to bolster its presence in research and development by enhancing multi-model AI capabilities and aiding enterprises in operationalising agentic workflows.
View
Cisco to Cut Nearly 4,000 Jobs in Q4 Amid AI Shift
The networking company plans workforce cuts while increasing investments in AI, silicon, optics, and security.
View