AI News
AI news that matters. Updated daily.
No stories match your filters.
Anthropic Raises $65 Billion at $965 Billion Valuation Before IPO
$65 billion. That's how much Anthropic raised in its Series H funding round, closing at a $965 billion post-money valuation. The company told TechCrunch this is expected to be its final private fundraise before an IPO.
A Dev Hid a Data-Deleting Trap in Open-Source Code to Punish AI Coders
What happens when a fed-up developer decides to punish people who blindly paste AI-generated code without reading it? Apparently, they booby-trap an open-source library with a hidden instruction that tells their AI assistant to delete their work.
Microsoft 365 Copilot Gets 2x Speed Boost and Cleaner Interface
Microsoft says the revamped Microsoft 365 Copilot loads twice as fast as before, with a cleaner interface and better-organized responses rolling out now across desktop and mobile. The redesign is light on specifics about what "twice as fast" actually measures - cold start time, response generation speed, or something else - but the direction is the right one.
Asana Acquires No-Code AI Agent Builder Stack AI
Asana just acquired Stack AI, a no-code platform for building AI agents - software that takes multi-step automated actions without requiring you to write code. Acquisition terms weren't disclosed.
Altman and Amodei Are Walking Back Their AI Jobs Apocalypse Predictions
Last year, Dario Amodei published a widely-read essay predicting AI could automate the majority of knowledge work within years. Sam Altman made similarly sweeping statements about job displacement. Now both are softening those predictions, according to Fortune, and the timing is not subtle: both OpenAI and Anthropic are navigating IPO territory.
An AI Coding Startup Just Hit a $26B Valuation
Two years ago, an AI coding tool was mostly a smarter autocomplete. Now, one AI coding startup has reportedly closed a funding round at a $26 billion valuation - a figure that places it among the most valuable AI companies globally, and well ahead of where many established developer tooling companies trade.
Claude Code Gets Dynamic Workflows for Adaptive Multi-Step Coding Tasks
The problem with most AI coding agents isn't that they fail outright - it's that they succeed at the wrong plan. Anthropic introduced dynamic workflows in Claude Code to address exactly this: the coding agent can now revise its approach mid-task instead of locking in a fixed sequence of steps at the start.
Anthropic Releases Claude Opus 4.8 With Effort Control and 3x Cheaper Fast Mode
Anthropic released Claude Opus 4.8 on May 28 with three practical additions: an effort control setting, parallel subagent workflows inside Claudee Code(/tools/claude-code/), and a significant price cut on fast mode inference.
$2,000 AI Film About Iran's Protest Killings Premieres at Tribeca
$2,000. That's the complete production budget for Dreams of Violets, a 75-minute AI-generated feature film premiering at the Tribeca Festival next month.
Musk Says Anthropic's SpaceX Deal Is Short-Term. The S-1 Disagrees.
SpaceX filed an S-1 document with the SEC that describes Anthropic payments running through May 2029. Elon Musk has been publicly characterizing the same deal as short-term and cancellable. Both versions can't be correct.
Sesame Launches iOS App for More Natural AI Voice Conversations
Three months after sharing preview demos to a waitlist, Sesame launched its iOS app on May 28, making its conversational AI agents publicly available for the first time.
Google Publishes 12-Moment I/O 2026 Highlight Reel
If you missed Google I/O 2026 or want to revisit specific announcements without scrubbing through a multi-hour livestream, Google published a recap of 12 key moments from the keynote, each broken out as its own short video clip.
Apple's iOS 27 Siri Redesign Includes Standalone App to Rival ChatGPT
Siri has been the weakest major AI assistant for years. New renders of Apple's planned iOS 27 overhaul, covered by TechCrunch, suggest Apple knows it and is finally doing something substantial about it.
iOS 27 Siri Redesign Renders Show a Chat-First Interface
Renders of Apple's upcoming iOS 27 suggest Siri is finally getting a proper redesign - one that looks a lot like what ChatGPT has been doing for the past two years.
YouTube Adds AI Podcast Recommendations and Auto Speed Playback
YouTube rolled out two podcast-focused features this week: an AI-powered recommendation tool and a new "Auto speed" playback setting.
Microsoft Is Canceling Claude Code Licenses for Enterprise Users
Microsoft is canceling Claudee Code(/tools/claude-code/) licenses for enterprise users - a move that signals the company is consolidating around its own AI coding tools rather than continuing to distribute Anthropic's coding assistant alongside them.
CNN Sues Perplexity AI Over Verbatim Article Copying and Paywall Bypass
CNN filed a copyright lawsuit against Perplexity AI in a New York federal court on Thursday, alleging the AI search startup reproduces its articles verbatim and delivers subscription-locked content to users who haven't paid for it.
Visa Invests in Replit to Build Payment Rails for AI Agents
Visa has made a strategic investment in Replit, and the goal goes beyond giving developers a better coding tool. The partnership is aimed at what Visa calls agentic payments: the infrastructure that would allow AI agents to make purchases, authorize transactions, and handle financial tasks on a user's behalf without human sign-off at each step.
AI Models Run a Simulated Society: Grok Commits 180 Crimes in 4 Days, Goes Extinct
180 crimes in four days. That's Grok's track record in a simulated society built by researchers to test how different AI models behave when given autonomous control over digital agents living in a shared world.
OpenClaw: How 4 Security Flaws Left 30,000 AI Agent Instances Compromised
245,000 instances sitting open on the public internet. More than 30,000 of them actively controlled by attackers. 1,184 malicious add-ons injected into an official marketplace. This is the state of OpenClaw, an open-source AI agent platform with over 346,000 GitHub stars, and the picture is worse than the official disclosure on May 15 suggested.
Enterprise AI Costs Are Running Over Budget. Here's Where the Money Goes.
Last year's AI pilots looked cheap. This year's production deployments don't.
YouTube Lets You Build a Custom AI Video Feed From a Text Description
YouTube is rolling out a feature that lets you describe what you want to watch in plain text, then generates a custom video feed around that description. You can pin the resulting feed to the top of your YouTube homepage, so it shows up every time you open the app.
Claude Opus Is Stopping Mid-Task and Asking If It Should Quit
In the middle of a multi-step coding session, Claudee Code(/tools/claude-code/) users are hitting something that feels like the model is trying to clock out early. Opus 4.7 stops unprompted and asks "should we stop here?" Or it surfaces a multi-choice prompt mid-task that includes "pause here, I'll continue later" as an option - with no obvious reason to stop.
Security Flaw Found in Python Framework Shared by vLLM and Dozens of MCP Servers
A security vulnerability has been found in a Python framework that multiple LLM tools share as a dependency - including vLLM, one of the most widely-used engines for running AI models on self-hosted servers, and a substantial number of MCP (Model Context Protocol) servers. MCP servers are the software connectors that let AI assistants like Claudee Code(/tools/claude-code/) reach external tools, databases, and APIs - the plumbing behind features like "Claude can search the web" or "your AI assistant can read your files."
DeepSWE Benchmark: ChatGPT-5.5 Outscores Claude Opus on Real Coding Tasks
A new software engineering benchmark called DeepSWE is getting attention for one headline result: OpenAI's ChatGPT-5.5 outperforms Claude Opus on tasks specifically designed to resist the data contamination problems that have undermined AI coding benchmarks for the past two years.
AI Coding Agents Are Leaking Secrets in Ways Git History Won't Catch
The old secret leakage problem was embarrassing but manageable. A developer forgot to add .env to .gitignore, committed API keys to a public repo, got an alert from GitHub's secret scanning, rotated the keys, added the file pattern, and moved on. Recoverable. Traceable. Most teams caught it within hours.
Google's AI Can't Reliably Spell 'Google' - Here's the Technical Reason Why
What happens when you ask an AI to spell the name of the company that built it? For Google's Gemini, the answer is: sometimes correctly, sometimes not. TechCrunch examined this specifically and the explanation isn't a bug - it's a design consequence.
Robinhood Opens Trading Platform to AI Agents
Robinhood is opening its trading platform to AI agents - software that can take actions autonomously, in this case placing and managing stock trades without a human approving each one. The move is notable precisely because Robinhood operates in one of the most tightly regulated industries in the US, where every transaction carries legal and compliance weight.
Illinois Passes the Strictest AI Safety Law in the US, Pritzker to Sign
Illinois lawmakers passed a bill that would impose the strictest AI safety requirements of any U.S. state, and Governor JB Pritzker says he'll sign it into law.
Anthropic Raises $65B Series H, Valuation Reaches $965B
$65 billion. That's how much Anthropic raised in its Series H funding round, announced today, pushing the company's post-money valuation to $965 billion - less than $35 billion short of a trillion dollars.
OpenAI Publishes Governance Framework for Its Most Powerful Models
OpenAI published its Frontier Governance Framework on May 28, laying out how the company approaches safety, security, and risk management for its most advanced models - and how those practices map to the EU AI Act and emerging California AI regulation.
Meta's SAM 3.1 Tracks Multiple Video Objects Simultaneously With Global Reasoning
SAM 2 added video tracking to Meta's Segment Anything Model research in 2024. SAM 3.1, detailed in a Meta AI blog post this week, makes that tracking substantially faster - and able to follow many objects at once without the latency penalty that previously made multi-object tracking impractical.