TechCrunch spoke to investors to find the hottest startups in the Spring 2026 YC batch. Some of them commanded valuations of over $175 million, VCs said.

Politico: Sources: the White House and Anthropic are working on a framework that would assess the severity of AI security flaws, a sign that negotiations are progressing  —  The attempt to create a standardized method to evaluate this and future such incidents underscores how the administration is racing …

Data scientists sit at the intersection of analytics, machine learning (ML) and AI,...

Jens Büttner/Getty Images A customer sends a WhatsApp message to a local store asking about a product. Within seconds, they receive an answer. The product they’re after is in stock. Or, if it isn’t, they’re recommended an alternative. Welcome to the next generation of customer service bots – and the latest battleground among the world’s largest technology companies. Big tech firms are now pouring billions of dollars into artificial intelligence (AI) agents: systems capable of autonomous decision-making and task execution. The value of agentic AI’s global market is projected to climb from US$10.9billion in 2026 to a staggering $182.9 billion by 2033. At its annual Conversations conference in London this month, Meta unveiled Business Agent, an AI system that can answer customer questions, qualify sales leads, manage bookings and process transactions directly within platforms such as WhatsApp. For small businesses that lack the staff or resources to provide round-the-clock customer support, the appeal is obvious. Larger organisations can integrate the technology into existing sales, booking and customer management systems. Having spent two decades reshaping the advertising industry, Meta’s move may seem like a natural progression. But there is more at stake than another tech giant automating customer service. Converting attention to transaction Most of us know Meta as the conglomerate behind social media platforms Facebook, WhatsApp, Messenger and Instagram. But Mark Zuckerberg’s business, which last year hit US$200 billion in revenue, has long been built around advertising. It comes down to understanding audiences, capturing their attention and then selling businesses the chance to reach the right people with the right message at the right time. Meta’s Business Agent now shifts the company into that transactional moment which follows the advert, click or message. It’s the point where a customer is already asking themselves what to buy, whether to book or how to solve a problem. This means Meta has begun seeking a place within the customer relationship itself: whether that’s answering questions, presenting options, organising follow-up or making a shopper’s next step easier to take. It also explains why AI agents have become such a hotly contested space for tech companies. The likes of Google, Amazon, Microsoft, OpenAI and Meta have all started from vastly different corners of the market but, increasingly, their ambitions are all converging here. Microsoft and Amazon Web Services are leveraging their massive cloud infrastructure to embed autonomous agents directly into existing enterprise resource planning and customer relationship management software, such as Dynamics 365. OpenAI is aggressively pushing custom-built, multi-agent frameworks that allow businesses to deploy tailored “GPTs” to handle complex, cross-department operations. Meanwhile, Google is integrating agentic capabilities directly into its dominant search and workspace tools, aiming to capture intent before a user even leaves the browser search bar. Meta’s route seems especially logical because many of those interactions already happen in its messaging platforms. Just consider a restaurant that can take bookings through WhatsApp or a fashion label able to handle product queries through Instagram. For Meta, its agent can convert those exchanges into a more fully automated commercial pathway. Who really wins? For small businesses, the benefits are obvious. AI agents offer capabilities once reserved for banks, airlines and other large organisations able to build sophisticated customer service systems of their own. They can answer routine questions, remember product details, respond across multiple languages and free staff to focus on tasks that still require human judgement, such as handling complaints. But convenience comes with a trade-off. The more useful an AI agent becomes, the more influence it gains over the interaction itself. It is helping determine what information customers receive, which products are recommended and how they move from enquiry to purchase. At the same time, every interaction provides the platform owner with valuable insight into what customers want, where they hesitate and what ultimately drives a sale. For many businesses, that may seem a fair exchange. But over time, the balance of power may begin to shift. As more customer interactions are mediated by AI, businesses risk becoming increasingly dependent on platforms they do not control. Customers, meanwhile, may enjoy faster responses and a more seamless experience. Less visible is the role these platforms play in shaping those interactions – and the commercial value they derive from them. Just how profoundly AI agents shake up global commerce remains to be seen. But the early signs suggest they will do more than automate customer service. They could accelerate a shift in power away from the businesses that own products and services, and towards the platforms that increasingly mediate the relationships those businesses depend on. Guy Bate is affiliated with EdTechNZ

The vulnerability, disclosed 12 months ago, affects multiple manufacturers.

Discussion | Link

Plaintiffs in the class -action complaint allege Rivian falsely promised for years it would bring hands-free driving to its first-generation R1 vehicles.

On today’s Uncanny Valley, we dive into the dysfunction in Meta’s newly formed AI unit and why it’s been driving already-low employee morale even further into the ground.

Yann LeCun said that he was "not very positive about the prospect of xAI." Bloomberg/Getty Images; Harun Ozalp/Anadolu via Getty Images Yann LeCun said that Elon Musk had trouble hiring top AI talent because of the xAI founding team's exits. The former Meta chief AI scientist called xAI "kind of a failure." LeCun and Musk have a long-running feud, including multiple heated X exchanges. Two of AI's biggest names have reignited their feud. Former Meta chief AI scientist Yann LeCun has long had a rocky relationship with Elon Musk. LeCun has bashed Musk's leadership strategies and his AI predictions. Musk has questioned LeCun's science. On Thursday, LeCun told CNBC that Musk's AI lab, xAI, was "kind of a failure." "Elon is now in a position where it's very, very difficult for him to hire top people in AI, because he's not behaved in very good ways toward the previous team," the AMI Labs founder said. LeCun was referring to xAI's founding team. All of the non-Musk cofounders have since left xAI. The final holdout, Ross Nordeen, was abruptly cut off from the company's systems and disappeared from a group chat, Business Insider previously reported. One of xAI's recent moves was to lease its data centers to competitors such as Anthropic. LeCun pointed to this as a sign of weakness. "He's got this huge infrastructure, which he rents to other people, because that's the only way he can recoup the costs," he said. LeCun said that he was "not very positive about the prospect of xAI." The interviewer asked whether LeCun thought xAI could compete on the frontier of AI. "No, I don't," he replied. XAI is now a part of SpaceX, which went public last week. SpaceX and xAI did not respond to requests for comment. In the interview, LeCun was more positive about Musk's other companies. "SpaceX is doing great stuff," he said. "I own a Tesla," LeCun said. He still took a jab: "Full self-driving is not full self-driving, but it's useful." Read the original article on Business Insider

Ben Weiss / Fortune: Sources: APEC, a derivatives exchange founded by the 22-year-old son of pro-crypto Senator Kirsten Gillibrand, raised $30M led by Lux at a $300M valuation  —  The 22-year-old son of a crypto-friendly senator plans to launch his own exchange for a type of derivative popularized by digital asset traders.

Discussion | Link

Discussion | Link

Discussion | Link

The threat group's curious business model may combine opportunistic monetization alongside intel collection, without much coordination between the two.

Discussion | Link

<p><strong>Release:</strong> <a href="https://github.com/datasette/datasette-acl/releases/tag/0.6a0">datasette-acl 0.6a0</a></p> <blockquote> <p>This release expands <code>datasette-acl</code> from table-only permissions toward a general resource-sharing system.</p> </blockquote> <p>Alex Garcia did most of the work for this release - we're fleshing out the plugin that will allow multi-user Datasette instances finely grained control over who can access which resources within Datasette.</p> <p>Tags: <a href="https://simonwillison.net/tags/datasette">datasette</a>, <a href="https://simonwillison.net/tags/alex-garcia">alex-garcia</a></p>

Artificial Analysis: GLM-5.2 is the leading open weights model on Artificial Analysis' Intelligence Index, scoring 51, only behind Fable 5's 60, Opus 4.8's 56, and GPT-5.5's 55  —  Z ai's GLM-5.2 is the new leading open weights model on the Artificial Analysis Intelligence Index scoring 51 and it sits on the Pareto frontier of Intelligence vs Cost per Task

About 47% of singles look negatively at the use of AI in dating -- but many dating app users are open to AI helping with profile punch-ups and conversation starters.

Discussion | Link

This week was not an exceptionally busy one for large funding deals, though we saw sizable rounds in a lively mix of sectors ranging from AI to fintech to quantum computing, biotech and cybersecurity.