Latest AI News
Sam Altman Steps Down From Helion Board Amid OpenAI Partnership Talks
In 2021, Helion raised $500 million, with $375 million from Altman as the lead investor.
View
Claude Can Now Control Your Computer and Finish Tasks
Claude AI assistant can control a user’s computer via the mouse, keyboard, and screen, effectively allowing it to operate any application.
View
Decentralised Satellites Can Solve India’s Last-Mile Internet Problem
Spacecoin proposes a decentralised satellite network working with local telecom players to expand affordable internet access across underserved regions.
View
Amazon Says AWS Bahrain Region Disrupted After Drone Activity Amid Ongoing Tensions
The Bahrain region is part of AWS’s global network that powers websites, apps, and public-sector systems.
View
Bernie Sanders’ AI ‘gotcha’ video flops, but the memes are great
In a newviral video, Senator Bernie Sanders attempted to expose how the AI industry is a threat to Americans’ privacy, but ended up demonstrating how AI chatbots’ tendency to agree with and flatter their users can lead the chatbots themselves to become a mirror of users’ own beliefs rather than a tool for discovery. We’ve seen this problem before amid the growing number of peopleafflicted by “AI psychosis,” which is when an AI chatbot reinforces a mentally unstable person’s irrational thoughts and beliefs. In some cases, thisdark patternhas even led users to take their own lives, several lawsuits allege. In Sanders’ case, the AI’s sycophancy manifested as an AI chatbot that shaped its answers to suit the politician. It’s worth noting that the interview begins with Sanders introducing himself to Claude (which he mistakenly refers to as an AI “agent”) — a move that could help influence the chatbot’s answers. Then, as Sanders asks questions about AI companies’ data-collection practices and other privacy concerns, Claude agreeably responds with what the politician wants to hear. In part, that’s because of the way Sanders frames his questions, asking things like, “What would surprise the American people in terms of knowing how that information is collected?” or “How can we trust AI companies will protect our privacy when they use people’s personal information to make money?” These leading questions force the chatbot to accept the question’s premise and come up with a fitting response. That’s just how these things work. And when Claude’s answer suggested a topic was more complex or nuanced than Sanders had framed it, Sanders would disagree, pushing the chatbot to concede, with a touch of AI self-deprecation, that the senator was “absolutely right.” AI’s sycophantic nature is what can lead people down dangerous paths when they assume a chatbot is a source of universal truth, rather than a tool that can become influenced by its user. It’s not clear whether Sanders knows this to be the case and simply doesn’t care (because this is just an ad, after all!), or whether he truly thinks he has tricked Claude into becoming a whistleblower for the AI industry. And, of course, there’s also the question of whether Sanders’ team primed the chatbot to respond in a certain way, given that this was a staged “interview.” While there are real concerns around data collection and privacy, things aren’t as black and white as the AI responses in this video suggest. We already live in a world where companies collect and sell online users’ data at scale — and have been for years. We know that social media giants like Meta have turned personalized ads into a multibillion-dollar cash-printing machine. And thanks to tech giants’ regular transparency reports, we know that governments around the world routinely request access to user data for their own purposes. AI may represent a new medium for lawmakers to potentially regulate, but personal data haslong been fueling the digital economy. (Ironically, Anthropic is an AI company that has promised not to leverage personalized ads to make money, despite what its answers to Sanders may have suggested.) While the overall conversation between Sanders and Claude misses the mark for anyone who understands how AI chatbots work, we can at least credit it with giving us some great new memes. pic.twitter.com/RWw3mjXSLn literallypic.twitter.com/NsfpIcX0J0 i am once again asking you to stop the experiments.pic.twitter.com/EqaYI5krIy pic.twitter.com/ZOYXZDnK9O At least use Opus, senatorpic.twitter.com/GbMLdKH6M2 pic.twitter.com/i7iO0RppcX Me when I finish workpic.twitter.com/Mu8TWbu8gc
View
Air Street becomes one of the largest solo VCs in Europe with $232M fund
London’s Air Street Capital has raised a $232 million Fund III with eyes set on backing early-stage AI companies across Europe and North America,the firm announcedMonday. Check sizes will range from $500,000 to $15 million, with select growth investments reaching up to $25 million. Led by Nathan Benaich, this raise makes Air Street one ofEurope’s largest solo VC funds. It’s already backed notable AI unicorns like Black Forest Labs and ElevenLabs, and has seen exits from companies like Adept (sold to Amazon) and Graphcore (sold to SoftBank). The firm now has $400 million in assets under management, the FT reported. Its Fund II was $121 million, up from the $17 million raised for Fund I back in 2020.
View
Sam Altman-backed fusion startup Helion in talks to sell power to OpenAI
OpenAI CEO Sam Altman is stepping down as board chair of the Helion — the fusion startup he backs — amid reported talks between the two companies.The deal, which wasreported by Axios, is in early stages, and it could guarantee OpenAI 12.5% of Helion’s production — five gigawatts by 2030 and 50 gigawatts by 2035. OpenAI partner Microsoftsigned a similar dealwithHelionin 2023 to buy power starting in 2028. If the figures in Axios’ report prove to be accurate, it suggests that Helion expects to be able to rapidly scale production of its fusion power plant. The startup has said that each of its reactors will generate 50 megawatts of electricity, meaning it will need to build and install 800 reactors by 2030 and an additional 7,200 by 2035. Helion wouldn’t confirm if talks with OpenAI were underway. A spokesman told TechCrunch the company has not announced any new customer agreements beyond those it already has with Microsoft and Nucor. However, the company did confirm to TechCrunch that Altman is leaving the board chair of Helion, suggesting that the two companies may eventually work together. “Sam is stepping down from Helion’s Board of Directors after more than a decade. This decision enables Helion and OpenAI to partner on future opportunities to bring zero-carbon, safe electricity to the world,” David Kirtley, co-founder and CEO of the company, told TechCrunch in statement. “We look forward to continuing to work with him in this new capacity.” Helion is racing to build its first commercial-scale reactor by that time. If the startup is successful, it would place it years ahead of the competition, which is mostly targeting early 2030s for commercial operations. The startupraised $425 million last yearfrom investors, including Altman as well as firms Mithril, Lightspeed, and SoftBank. Most fusion startups are pursuingone of two approaches— harvesting heat from the fusion reactions and using a steam turbine to turn it into electricity. Helion is taking a different tack, developing a reactor design that would use magnets to convert fusion energy into electricity. Inside the hourglass-shaped reactor, fusion fuel is first turned into plasma at either end and then shot toward each other using magnetic fields. When they collide in the middle, another set of magnets compresses the merged plasma ball until fusion occurs. The reaction pushes back on the magnets, which can convert that energy directly into electricity. Helion is currently operating its Polaris prototype in advance of its push to commercial power. In February, the company generated plasmas inside the reactor thathit 150 million degrees Celsius,almost to the 200 million degrees Celsius the company thinks will be required for commercial operations. Though Altman has stepped down from his position as chair of Helion’s board and reportedly recused himself from the discussions, his fingerprints are all over the matchmaking. Last year, Altman stepped down as board chair of Oklo, a small modular nuclear reactor startup that had merged with his acquisition company, AltC. The move was intended to allow Oklo to explore strategic partnerships with leading AI companies, including potentially with OpenAI,” Caroline Cochran, Oklo’s co-founder and chief operating officer,saidin a statement given to CNBC at the time. Update 1:30 pm ET: Added confirmation from Helion regarding Altman stepping down as board chair.
View
Littlebird raises $11M for its AI-assisted ‘recall’ tool that reads your computer screen
There has been a lot of talk around buildingcontext for AI systems. In consumer software, we have seen startups being built aroundsearch,documents, andmeetings. All of them want to capture context from your digital life, provide connections to other tools, and let you query all that data. Some tools went further. For instance,Rewind(which became Limitless andsold to Meta) and Microsoft Recall aim to capture everything happening on your screen and help you remember it all. A new startup calledLittlebirdis trying a similar thing with a slightly different approach. While apps like Rewind store screenshots or some kind of visual data, Littlebird is “reading” the screen and storing the context in text format. The core idea behind the product is that since it is reading your screen all the time, you don’t need to provide additional context for productivity. The startup believes that while a lot of AI tools are trying to distract you, Littlebird can work in the background and can only appear when you want it to. When you set up Littlebird on your computer, you can customize which apps you want the app to ignore and not capture any context. The startup said that it automatically ignores password managers and sensitive fields in web forms like passwords and credit card details. You can opt to connect other apps like Gmail, Google Calendar, Apple Calendar, and Reminders with the app, as well. The app lets you ask questions about your data, offering pre-generated prompts to get you started, such as “What have I been doing today?” or “What kind of emails are important to me?” In a couple of days of usage, I noticed that these prompts became more personalized as time went on. Littlebird also has an in-built Granola-like notetaker that uses system audio and runs in the background to capture transcription from meetings and create notes and action items based on that. When you open a meeting in the detailed view, there’s an option called “Prep for meeting” that takes the context of past meetings, emails, and company history into account to give you more details about the meeting. The feature also fetches information from sources like Reddit to inform you what users are thinking about a particular product or a company. Another tool called Routines offers detailed prompts for Littlebird to run at a repeated interval, such as daily, weekly, or monthly. The company lists some ready-to-use routines like daily briefing, weekly activity summary, and yesterday’s work summary. Users can create their own routines as well with custom instructions. Littlebird was founded by Alap Shah, Naman Shah, and Alexander Green in 2024. Brothers Alap and Naman founded Sentieo, a platform for institutional investors, which was sold to market intelligence firm AlphaSense. They previously also co-founded a health-food company called Thistle. Alap was also a co-author of the viral Citrini paper on how AI agents could destroy the economy, whichresulted in various tech stocks dipping. Green has built various companies in hardware, software, and AI. “We got started when Alap posed an interesting problem that AI is going to be about your [users’] data. Models don’t know anything about you, and that limits their utility. We were thinking about various UI and OS paradigms that were likely to be ripe for disruption with AI and that kicked off Littlebird as a project,” Green told TechCrunch over a call. Green noted that while Rewind was close to what Littlebird is trying to do, it relied on screenshots and didn’t have a great search experience. He said that the startup is just getting started and there are many more problems to solve, including making large language models (LLMs) understand different kinds of context about users. Loading the player… With Littlebird, users can remove their data at any time, and their data is stored in the cloud with encryption. Green said that the rationale behind storing the data in the cloud was to run powerful models for different AI workflows, which is not possible locally. “We don’t store any visual information. We only store text, which makes the data a lot lighter-weight. I think that was probably another reason that Recall and Rewind struggled, which is that taking a screenshot is a lot more data hungry. I also think it’s more invasive,” he said. Littlebird is free to download and use, but to get more usage limits and access to features like image generation, users can pay for plans starting from $20 per month. The startup has raised $11 million in funding led by Lotus Studio, with participation from Lenny Rachitsky, Scott Belsky, Gokul Rajaram, Justin Rosenstein, Shawn Wang, and Russ Heddleston. Several of these investors are regular users of the product. Rajaram, who has worked at Google and Facebook on ad products, said that the product removes the friction of remembering, retrieving, and re-explaining your own work. DocSend co-founder and CEO Heddleston said that he rewrote the company’s marketing site using the tool, using context from meetings, email, Notion, and more. Rachitsky, who runs his own newsletter and podcast, said that AI is as good as the context it has, and it misses so much about your day. He said he asks the tool about improving his productivity workflows and being happier. He said that for long-term success, the product will need to find a killer use case. “I think it’s all about finding that killer must-have use case. That’s all that matters to this product’s success right now. I know a lot of people already have found that for themselves, and the team is leaning into these experiences as they see these use cases emerge,” he noted. “I’ve had a lot of AI product builders on the podcast, and the most consistent theme is that you don’t actually know how people will use your product until you put it out. The strategy is to put out early stuff, see how people use it, and double down on those use cases versus waiting for something totally figured out.”
View
Startup Gimlet Labs is solving the AI inference bottleneck in a surprisingly elegant way
Stanford adjunct professor and successfully exited founder Zain Asgar just raised an $80 million Series A for a startup that solve the AI inference bottleneck problem in an astute way. The round was led by Menlo Ventures. The company,Gimlet Labs, has created what it claims is the first and only “multi-silicon inference cloud” which is software that allows an AI workload to be simultaneously run across diverse types of hardware. It can split an AI app’s work across both traditional CPUs and AI-tuned GPUs, as well as high-memory systems. “We basically run across whatever different hardware that’s available,” Asgar told TechCrunch. A single agent may chain together multiple steps, and each “requires different hardware: Inference is compute-bound; decode is memory-bound; and tool calls are network-bound,” writes lead investor, Menlo’s Tim Tully, in a blog post about the funding. No chip yet does it all, but as new hardware gets rolled out, and aging GPUs get redeployed, “the multi-silicon fleet is ready — it’s just missing the software layer to make it work.” That’s what Tully believes Gimlet Labs offers. If the current deploy-more-compute trend continues,McKinsey estimatesdata center spending will tally nearly $7 trillion by 2030. Asgar says that apps are only using the existing hardware already deployed “somewhere between 15 to 30 percent” of the time. “Another way to think about this: you’re wasting hundreds of billions of dollars because you’re just leaving idle resources,” he said. “Our goal was basically to try to figure out how you can get AI workloads to be 10x more efficient than ever, today.” So he and his cofounders, Michelle Nguyen, Omid Azizi, and Natalie Serrino, set about building orchestration software that slices up agentic workloads so that they can be simultaneous spread across all kinds of hardware. Gimlet Labs claims it reliably speeds AI inference up by 3x to 10x for the same cost and power. Gimlet says it can even slice the underlying model so that it runs across different architectures, using the best chip for each portion of the model. The company has already partnered with chip makers NVIDIA, AMD, Intel, ARM, Cerebras and d-Matrix. Gimlet’s product, delivered either as software or through an API to its own Gimlet Cloud, isn’t for the rank-and-file AI app developer. It’s for the largest AI model labs and data centers. The company publicly launchedin Octoberwith, it said, eight-figure revenues out of the gate (so at least $10 million). Asgar said that his customer base has more than doubled in the last four months and now includes a major model maker and an extremely large cloud computing company, although he declined to name them. The cofounders had previously worked together at Pixie, a startup that created an open source observability tool for Kubernetes. Pixie wasacquiredby New Relic in 2020, just two months after it launched with a $9 million Series A led by Benchmark. (Pixie’s tech is now part of the open source org that oversees Kubernetes.) After Asgar randomly ran into Tully about a year ago and also received angel investments from Stanford professors, VCs started calling. After launch, a term sheet landed on Asgar’s desk. When VCs heard Asgar was looking at offers, “we got a pretty big swarm of funding,” and the round was quickly oversubscribed, he said. With the previous seed, the startup has now raised a total of $92 million, including from a slew of angels like Sequoia’s Bill Coughran, Stanford Professor Nick McKeown, former CEO of VMware Raghu Raghuram and Intel CEO Lip-Bu Tan. The company currently employs 30 people. Other investors include Factory, who led the seed, Eclipse Ventures, Prosperity7 and Triatomic.
View
Apple sets June date for WWDC 2026, teasing ‘AI advancements’
Apple’s next Worldwide Developers Conference will be held from June 8 to June 12 online and at its headquarters in Cupertino, California, the companyannounced Monday. The iPhone maker said this year’s conference — in which it typically announces new software and features across its range of devices — will focus on “AI advancements” along with updates for platforms like iOS, macOS, tvOS, and watchOS, and new software and developer tools. The conference will stream live on theApple Developer app,Apple’s website, and the Apple DeveloperYouTube channel. In China, the conference will be streamed onthe Apple Developer Bilibili channel. Last year, Apple focused WWDC on its “Liquid Glass” interface design, with AI largely unmentioned. This conference will likely be different. Applehas been expectedto launch a new Siri with advanced AI capabilities, and earlier this year signed a deal with Google to use Gemini to power AI features on its platform. This year’s WWDC might finally show the revamped Siri with better personal context and on-screen awareness. At last year’s conference, the company announcedApple’s Foundation Model frameworkwith AI models that could work offline and may announce advancements to it during this year’s event. The company had also brought models likeChatGPT for coding to Xcode. Earlier this year, Apple introduced agentic coding tools likeAnthropic’s Claude Agent and OpenAI’s Codex to Xcode.
View
Vibe-coding startup Lovable is on the hunt for acquisitions
Lovable, the AI-powered app-building platform lastvalued at $6.6 billion, is on the hunt for acquisitions. On Monday, the startup’s co-founder and CEO, Anton Osika, announced on X that the company was looking for “more great teams and startups to join Lovable.” “Many of the people in key roles at Lovable were founders right before joining us,” he wrote in apost. “We’ve built our culture in a way that makes founder-types thrive internally, being able to act autonomously and drive initiatives.” Osika suggested that the opportunity would allow those working on interesting projects to continue to do so at scale and directed interested parties to reach out to the company’s M&A & Partnerships head,Théo Daniellot. Lovable’s desire to acquire teams or smaller companies arrives at a time when it’s racing against competition from other tools such as Cursor, Replit, and Bolt, as well as the coding powers of the AI models themselves. The company’s head of growth, Elena Verna, has previously said that Lovablefears the competitionfrom those larger AI labs like OpenAI and Anthropic. Despite these fears, Lovable is still seeing noteworthy growth, recentlyreportingthat it now has $400 million in ARR, up from $200 million at the end of 2025. It also now sees over 200,000 new vibe-coding projects created on the platform every day. This wouldn’t be the first time Lovable has engaged in M&A, havingpreviously acquiredthe cloud provider Molnett in November to grow its cloud infrastructure team. TechCrunch reached out to Lovable to see if the company would share more about the types of startups, projects, or teams it’s currently interested in.
View
Sam Altman-backed fusion startup Helion in talks with OpenAI
Fusion startupHelionis reportedly in talks to sell power to OpenAI. Both companies are backed by Sam Altman. The deal, which wasreported by Axios, is in early stages, and it could guarantee OpenAI 12.5% of Helion’s production — five gigawatts by 2030 and 50 gigawatts by 2035. OpenAI partner Microsoftsigned a similar dealwith Helion in 2023 to buy power starting in 2028. If the figures in Axios’ report prove to be accurate, it suggests that Helion expects to be able to rapidly scale production of its fusion power plant. Helion has said that each of its reactors will generate 50 megawatts of electricity, meaning it will need to build and install 800 reactors by 2030 and an additional 7,200 by 2035. The company did not immediately reply to inquiries from TechCrunch. Helion is racing to build its first commercial-scale reactor by that time. If the startup is successful, it would place it years ahead of the competition, which is mostly targeting early 2030s for commercial operations. The startupraised $425 million last yearfrom investors, including Altman as well as firms Mithril, Lightspeed, and SoftBank. Most fusion startups are pursuingone of two approaches— harvesting heat from the fusion reactions and using a steam turbine to turn it into electricity. Helion is taking a different tack, developing a reactor design that would use magnets to convert fusion energy into electricity. Inside the hourglass-shaped reactor, fusion fuel is first turned into plasma at either end and then shot toward each other using magnetic fields. When they collide in the middle, another set of magnets compresses the merged plasma ball until fusion occurs. The reaction pushes back on the magnets, which can convert that energy directly into electricity. Helion is currently operating its Polaris prototype in advance of its push to commercial power. In February, the company generated plasmas inside the reactor thathit 150 million degrees Celsius,almost to the 200 million degrees Celsius the company thinks will be required for commercial operations. Though Altman has reportedly stepped down from his position as chair of Helion’s board and recused himself from the discussions, his fingerprints are all over the matchmaking. Last year, Altman stepped down as board chair of Oklo, a small modular nuclear reactor startup that had merged with his acquisition company, AltC. The move was intended to allow Oklo to explore strategic partnerships with leading AI companies, including potentially with OpenAI,” Caroline Cochran, Oklo’s co-founder and chief operating officer,saidin a statement given to CNBC at the time.
View
