Latest AI News
The AI world is getting ‘loopy’
On Friday, Claude Code creator Boris Cherny made an appearance at Meta’s @Scale conference and, surprisingly, the first question from the audience was about loops. “Are loops the next hype cycle,” the questioner asked, “or are they for real?” Cherny’s answer was an emphatic, “Yes, they’re for real.” “Two years ago, we wrote source code by hand. We started to transition so agents write the code. And now we’re transitioning to the point where agents are prompting agents that then write the code,” he continued. “As big as the step from source code to agents was, loops are just as important and as big a step.” Later in the talk (around the 32:00 mark in the YouTube video posted above), Cherny got specific about the loops he keeps running in his own work. One agent is continually looking for ways to improve the code architecture, while another looks for duplicated abstractions that can be unified. They submit pull requests like any other coder, and since the code is constantly changing, they never stop running. It’s a powerful idea, particularly with a figure as significant as Cherny behind it. With the shift to agentic AI, the focus for most users has been managing their agents as well as possible: establish clear goals, check in on discrete units of progress, and don’t let them stray too far beyond the prompt. The loop takes it a step further by authorizing a swarm of agents to work continuously in the background, endlessly. It’s a lot of trust to place in AI — but with models getting better fast, it could be the next step in getting AI to handle real work. The first thing to recognize is that this isn’t entirely new. Recursive loops — functions that call themselves in order to repeat an action, along with a condition that stops the loop — are a mainstay of intro computer science courses. These loops are following a non-deterministic logic — that is, it’s a subagent that chooses when to stop the loop instead of a clear condition — but the same basic approach is at work. As soon as programmers started using AI to complete tasks, some version of the recursive loop, with AI overseeing AI, was bound to come up. Unlike classic computing, agentic loops can be maddeningly simple. One of the most popular tricks isthe Ralph Loop(named for Ralph Wiggum), which basically sums up all the work that the model has done and asks if it’s accomplished its goal. It’s a way of dealing with AI models getting lost as they run for too long — essentially bouncing the model back and forth until the task is complete. Another way to think of loops is as part of the general push for more test-time compute. As OpenAI researcher Noam Brown observedearlier this month, contemporary models can solve nearly any problem if you throw enough compute at them. That means one way to ensure a problem gets solved is to just keep throwing compute at it until it’s finished. That’s particularly true for hill-climbing problems like improving a code base, where the model can just keep making incremental improvements until it reaches a given threshold. Or, as in Cherny’s example, it can just keep making incremental improvements for as long as there’s compute to spend on it. If that sounds expensive, it should. Like agentic AI before it, AI loops burn through tokens a lot faster than simple Q&A chatbots — and because the point is to keep the loop running all the time, there’s no ceiling to how much you can spend. That’s fine for Anthropic, which is ultimately in the token-selling business, but for everyone else, it may be a pricey way to work. Still, depending on the problem the agentic loop is trying to solve, and the right setup that allows for oversight of token spend, drift, and other classic AI issues, the benefits could be staggering enough to outweigh the costs.
View
SpaceX inks compute deal with Reflection AI, an open source AI lab
First cameAnthropic, thenGoogle. Now, open source AI startup Reflection is tapping SpaceX for its abundant source of AI chips. Reflection AI will pay $150 million a month beginning July 1, 2026 through 2029 for immediate access to Nvidia’s latest GB300 AI chips and supporting hardware across SpaceX’s Colossus 2 data center near Memphis, Tennessee, the company told TechCrunch. The deal is worth up to $6.3 billion and either company has the option to end the contract with 90 days’ notice after the first three months. The deal is smaller than SpaceX’s deals with Anthropic and Google, which cost the companies $1.25 billion per month and $920 million per month, respectively. Those contracts also run through July 2029, although Elon Musk haspublicly downplayedthe three-year term, emphasizing that the contracts can be canceled at any time. Reflection used the compute deal — its first — to tout the value ofits open-weight AI strategy, which it has pitched as an open source alternative to closed frontier labs like Anthropic and OpenAI. Open-weight AI models, which publicly release their trained parameters, have received more attention following theU.S. government’s banof Anthropic’s closed models, Fable and Mythos. Thestartup, which was founded in 2024 by two former Google DeepMind researchers, said the compute deal is one of the largest announced open AI infrastructure commitments to date. “Recent events highlight how important open source is to the AI ecosystem, with more nations and enterprises recognizing the risks and costs associated with exclusively depending on closed models,” a spokesperson said in an emailed statement. “Our deal with SpaceXAI signals Reflection’s strategic importance within the frontier AI ecosystem, and more compute means more runway to build the world’s best open models at scale.” TheColossus data centerwas originally built by xAI, a company founded by Elon Musk that is now part pf SpaceX, for its own AI efforts. As its internal pursuits have faltered, SpaceX leveraged its valuable AI chip holdings and began renting them out to some of the world’s top AI labs.
View
Amazon is testing Alexa+ in India with Hindi support
Amazon is planning to increase the footprint of its new conversational AI assistant Alexa+ to India and is inviting users in the country to test out a Hindi-language version. The company sent out emails to some customers, which were seen by TechCrunch, asking users to fill out a form in Hindi by June 22 to join the beta-testing program. “You are invited to join the Alexa+ Beta programme in India. We are creating a new Alexa experience, and your feedback will be important to refine what Alexa+ will be able to do. By joining the Alexa+ Beta programme, you’ll be notified when the testing experience in Hindi (India) is available to you,” the email read. The email added that the beta software would have bugs and might give inaccurate information or mispronounce local nuances. The company confirmed that it is testing Alexa+ in India but didn’t provide a comment. At the moment, Alexa+ is not available in India, and it is not clear when it will launch in the country. Amazon launched Alexa in India with English support in 2017 and addedHindi compatibility in 2019. More than 600 million people speak Hindi in India, and Amazon is trying to tap the market of native speakers who might speak both Hindi and English in a code-mixed way. Companies know that voice might be a big factor in AI tool usage in India and are findingnew ways to have people talk to assistants. Amazon first announcedthe generative AI-powered conversational assistant Alexa+ in 2025. However, its rollout was slow, and the new experience was madeavailable to all U.S. users only in February. This year, the company has increased Alexa+’s footprint to countries like the U.K., Canada, Brazil, Mexico, Italy, and Germany, with support for local context. Amazon offers Alexa+ to Prime customers for free, and others can pay a monthly fee to access the updated assistant.
View
Micron Invests in Anthropic, Partners on AI Memory Infrastructure
Micron and Anthropic will work together on memory and storage technologies used in AI training and inference
View
The founder conference built for growth: TechCrunch Founder Summit pass rates increase June 26
Founders don’t grow alone. They learn from peers tackling similar challenges, connect with those who have already navigated the road ahead, and build relationships with investors aligned with their vision. For the next five days, save up to $190 on your pass toTechCrunch Founder Summit 2026. On November 4 in Boston, TechCrunch Founder Summit 2026 brings together 1,000+ founders and investors for a day focused on growth, execution, and building lasting connections. This is TechCrunch’s flagship founder conference, designed for founders first. The experience centers on peer-to-peer learning, candid conversations, and practical, expert-led insights, with opportunities to connect directly with investors aligned with your goals and stage of growth. Early Bird pricing is available through June 26 at 11:59 p.m. PT. Save up to $190 on your pass or save even more with group discounts of up to 30% for teams of four or more.Register here. TechCrunch Founder Summitis built to help founders accelerate growth by learning directly from experienced operators, successful founders, and investors who understand the realities of building a company. At Founder Summit, you’ll connect with: Whether you’re preparing to raise capital, refining your pitch, or working toward your next milestone, Founder Summit creates opportunities for meaningful conversations and high-value connections.Grab your Early Bird ticket savings before June 26. Founder Summitsessions focus on the moments that shape a company’s trajectory. Breakout and roundtable discussions are designed to be actionable, relevant, and immediately useful. Past topics have included: These sessions are designed to give founders the clarity, confidence, and strategies needed to reach the next stage.Register nowto lock in up to $190 in ticket savings. Past speakers have shared candid lessons on building companies, raising capital, and navigating growth, including: Additional speakers have included leaders from Sequoia Capital, NFX, Underscore VC, Glasswing Ventures, Wing Venture Capital, Construct Capital, Greylock, and Precursor Ventures. The 2026 agenda is taking shape, with additional founders, operators, and investors to be announced soon.Visit the event websitefor the latest updates. If you’re interested in leading the conversation in a roundtable or breakout session,submit your speaking topicfor a chance to be voted onto the agenda by the TechCrunch audience. TechCrunch Founder Summit 2026is where founders come to learn faster, connect more intentionally, and gain practical insights for growth. Join founder-focused conversations. Learn proven strategies. Build relationships that support your next stage of growth. Save up to $190 on your pass and up to 30% when registering as a group by June 26 at 11:59 p.m. PT.Register now and secure your spot in Boston. Interested in exhibiting at TechCrunch Founder Summit 2026?Reserve your exhibit tableand connect directly with founders, investors, and startup decision-makers.
View
How an Indian Pipe Maker is Quietly Building AI-Driven Factory Floor
Ashirvad by Aliaxis, the Bengaluru-based pipe manufacturer, is deploying AI across predictive maintenance, demand forecasting, and quality control.
View
BHASHINI, Assam Innovation & Startup Foundation to Drive Multilingual AI Innovation
The two organisations will also work on collecting, curating, and enriching linguistic datasets through the Bhashadaan platform
View
The Indian Drone Industry’s Shift Away From Manufacturing to Software
While hardware remains important, industry leaders argue that the real competitive advantage lies in the software that drives these systems.
View
Meta Invests $900 Mn in CRED, Hires Kunal Shah to Lead WhatsApp
CRED’s Chief Strategy and Finance Executive, Miten Sampat, will take over as interim CEO.
View
UPI Runs on a 100% Open-Source Stack. Now, NPCI Is Open-Sourcing Its Own Innovations
“We launched UPI in 2016 and reached our first billion transactions by 2019. In a very short time, we scaled massively into a national payment system."
View
BharatGen Joins AI Alliance’s Project Tapestry to Build Sovereign AI
The initiative seeks to bring together data, compute resources, talent, and funding from multiple countries and organisations to develop frontier AI systems.
View
Cyient Foundation Partners with Telangana Universities, Govt to Build Future Skills Ecosystem
Cyient Foundation has proposed an investment of approximately ₹7.5 crore over three years.
View
