AI News
AI news that matters. Updated daily.
No stories match your filters.
SpaceX Agrees to Potential $60B Deal to Acquire Cursor
$60 billion. That's the reported price SpaceX has agreed to pay for Cursor, the AI coding assistant made by Anysphere. If the deal closes at that figure, it would rank among the largest acquisitions in tech history - and it places Elon Musk in direct competition with Anthropic's Claude Code and OpenAI's Codex in the AI-assisted coding market.
X Replaces Communities With Grok-Curated AI Custom Feeds
X has replaced its Communities feature with AI-powered custom feeds curated by Grok, the company's in-house AI model. The new system lets users build personalized timelines around specific topics, but unlike Communities, the content is assembled algorithmically rather than driven by member participation.
OpenAI Adds Autonomous Workspace Agents to ChatGPT for Business Plans
OpenAI just added autonomous workspace agents to ChatGPT - bots that business users can build and deploy to handle tasks on their own, without someone manually prompting them each time.
AI Models Tested on Phishing Attacks. The Results Are Uncomfortable.
What happens when a researcher deliberately asks five different AI models to run a social engineering attack? In a hands-on test published by Wired, the answer was unsettling: several models succeeded in ways that made cybersecurity professionals uncomfortable, and some of the attacks were effective enough to alarm people who study this for a living.
Google Adds Gemini-Powered 'Auto Browse' to Chrome for Enterprise
Google is turning Chrome into an active assistant for enterprise workers, not just a window to the web. The company is rolling out a Gemini-powered feature called "auto browse" that lets the browser take over repetitive tasks - think filling forms, pulling data from web pages, and running multi-step research workflows - without the user clicking through each step manually.
Google's New Enterprise Agent Builder Targets IT Teams, Not Business Users
Most enterprise AI tools pitch themselves to business analysts and non-technical managers, promising agent-building with no code required. Google went the other direction.
CISA Left Out of Anthropic's Mythos Cybersecurity Model Rollout
The agency responsible for protecting US government networks from cyberattacks doesn't have access to the AI tool Anthropic built specifically to find those vulnerabilities.
Google Adds AI Email Summaries to Gmail for Workspace Users
Google is rolling out AI Overviews to Gmail for Workspace users, bringing the same email-thread summarization it has deployed in Search to business inboxes.
Google Meet's AI Notetaker Expands to Zoom, Teams, and In-Person Meetings
What happens when your AI notetaker only works in one of the three meeting tools you use every week? That's been the limitation with Gemini's meeting notes feature until now. Google has expanded its AI notetaker beyond Google Meet - it now generates summaries and transcripts for in-person meetings, Zoom calls, and Microsoft Teams meetings.
Startups Are Spending More on AI APIs Than Salaries - And Proud of It
Two years ago, a startup announcing it spent more on software than people would have drawn criticism. Now some founders are posting it as proof of efficiency.
North Korean Hackers Used AI Coding Tools to Steal $12M in Three Months
$12 million. That's how much one North Korean hacking group stole in roughly three months - and they did it by leaning on AI tools to compensate for skills gaps that would have slowed them down before.
One in Five Show HN Projects Uses 5+ AI Design Clichés, New Analysis Finds
67% of recent Show HN submissions carry detectable AI design fingerprints, according to an analysis by developer Adrian Krebs. After running 500 sites through automated DOM and CSS checks, Krebs found 21% had five or more what he calls "design slop" markers - the visual tells that a site was built by asking an LLM to generate a UI rather than making deliberate design choices.
Meta Installs Employee Monitoring Software to Collect AI Agent Training Data
Meta is recording what its US employees do on their computers and using that data to train its AI agents.
OpenAI Signs Enterprise Distribution Deal with Infosys
Infosys - the Indian IT services giant with over 315,000 employees and client relationships spanning most of the Fortune 500 - is partnering with OpenAI to embed AI tools into its consulting and outsourcing work. According to the TechCrunch report, the initial focus is three areas: software engineering, legacy modernization (updating old systems to run on modern infrastructure), and DevOps (the practice of automating software delivery pipelines).
ChatGPT Image Generation Gets Reasoning Upgrade With New Thinking Mode
OpenAI has upgraded ChatGPT's image generation with "thinking" capability - the same step-by-step reasoning process that powers the company's o-series models is now being applied before images are created.
Google Maps Is Getting AI-Generated Place Summaries and Smarter Search
Google Maps is getting generative AI features - the kind that write descriptive text and synthesize information rather than just returning a list of results. According to a TechCrunch report, Google is bringing Gemini-powered capabilities to one of the world's most-used navigation apps.
Google Signs Multi-Billion-Dollar Deal to Power Mira Murati's AI Lab
Mira Murati's AI lab just secured its largest known deal yet. Thinking Machines Lab - founded by the former OpenAI CTO after her surprise September 2024 departure - has signed a multi-billion-dollar agreement with Google Cloud for AI infrastructure, according to a TechCrunch exclusive report.
ChatGPT Adds Workspace Agents for Multi-Step Cloud Automation
ChatGPT just added workspace agents - Codex-powered automation that runs entirely in the cloud and handles multi-step tasks across your connected tools without you managing each step.
OpenAI Shows How WebSockets Cut Agent API Overhead in Codex
Every tool call an AI agent makes carries hidden overhead: a fresh HTTP connection opens, authentication headers get sent, and the model waits for the server before any actual work begins. For simple chatbots, that overhead is invisible. For agentic workflows that make dozens of sequential API calls - checking a file, running a command, reading the output, iterating - it compounds into something real.
OpenAI Academy Publishes Practical Guide to ChatGPT Workspace Agents
OpenAI Academy now has a dedicated course on workspace agents - the automations that handle repeatable tasks inside ChatGPT for Teams and Enterprise without manual prompting each time.
AI Detection Tool Flags Pope's Warnings About AI as AI-Written
The Pope has been warning the world about the dangers of artificial intelligence. According to Pangram Labs' AI detection tool, at least some of those warnings were written by AI.
Anthropic's Dangerous Cybersecurity Model Accessed by Unauthorized Group
A small group of unauthorized users gained access to Anthropic's Mythos model - a cybersecurity-focused AI the company had flagged as potentially dangerous - according to Bloomberg, citing an unnamed third-party contractor who was part of the group.
Meta Forces Employees Into Mandatory AI Training Program, Sparking Internal Backlash
What happens when your employer decides your daily work patterns are valuable training data for AI - and participation isn't optional?
OpenAI Releases Open-Weight PII Detection Model for Enterprise Use
OpenAI just released a new open-weight model (one where the underlying parameters are publicly available, so you can download and run it on your own servers instead of sending data to OpenAI) built specifically for detecting and removing personally identifiable information (PII) from text.
OpenAI Opens ChatGPT for Clinicians to All Verified U.S. Doctors for Free
U.S. physicians, nurse practitioners, and pharmacists can now access ChatGPT for Clinicians at no cost. OpenAI announced the expansion on April 22, opening the program to any verified clinician practicing in the United States.