⚡ Anthropic raises $16.8B ahead of IPO

Anthropic hit $16.8B in revenue ahead of its IPO while granting safety researchers unprecedented employee-level access, and Meta launched a powerful new open-weight AI model for Macs as DHS banned DeepSeek from all federal civilian agencies.

AI-generated summary of the last 7 days, updated daily

Ex-OpenAI Chief Launches Drug Chemistry Startup with $215M

Ex-OpenAI Chief Launches Drug Chemistry Startup with $215M

Ex-OpenAI Chief Launches Drug Chemistry Startup with $215M - Kevin Weil, who left OpenAI six months ago, has launched Warp Sciences with $215 million in funding. The startup is building a U.S.-based chemistry contract research organization to supply biotech and pharmaceutical companies with drug candidate molecules. Warp's first service is lead optimization a...

AI STARTUP FUNDING
Nirva Raises $33M for AI Jewelry Pendant

Nirva Raises $33M for AI Jewelry Pendant

Nirva Raises $33M for AI Jewelry Pendant - Nirva, founded in 2025 by two former Meta Reality Labs employees and a fashion entrepreneur, has raised $33 million total to build a jewelry-like AI pendant. The device weighs eight grams, lasts three days on a charge, and passively listens only to the wearer by filtering out surrounding voices. It ...

AI STARTUP FUNDING
Nirva raises $25M for AI emotional insight jewelry

Nirva raises $25M for AI emotional insight jewelry

Nirva raises $25M for AI emotional insight jewelry - Nirva Labs has raised $25 million, bringing total funding to $33 million, to develop AI-powered jewelry that passively listens to users' days and surfaces emotional patterns and relationship insights via a companion app. Backed by Lightspeed Venture Partners and South Park Commons, the company's nec...

AI STARTUP FUNDING
Agentic AI Foundation expands to ten open-source projects

Agentic AI Foundation expands to ten open-source projects

Agentic AI Foundation expands to ten open-source projects - The Agentic AI Foundation has added four open-source AI projects, bringing its total to ten since launching under the Linux Foundation last December. OpenAI donated Open Responses, a standardized template for AI model APIs that lets developers move code between providers without major rewrites. Anth...

AI OPEN SOURCE TOOLS
Action State Launches Open-Source AI Agent Verification Tools

Action State Launches Open-Source AI Agent Verification Tools

Action State Launches Open-Source AI Agent Verification Tools - Action State Group has released open-source tools to help businesses independently verify what AI agents did during cross-company transactions. The suite includes EvidenceBook, which stores tamper-evident signed transaction records; Terms Packs, which evaluate agent behavior against regulatory or bu...

AI STARTUP LAUNCH
Ai2 Launches OlmoEarth v1.3 Satellite Embedding Model

Ai2 Launches OlmoEarth v1.3 Satellite Embedding Model

Ai2 Launches OlmoEarth v1.3 Satellite Embedding Model - Ai2 released OlmoEarth v1.3, a satellite map embedding model that compresses a year of satellite observations into numerical fingerprints for 10-by-10-meter ground patches, with larger resolutions also available. The model draws from Sentinel-1, Sentinel-2, and Landsat imagery to help identify land ...

NEW AI MODEL LAUNCH
Mave Raises $34M for Agentic Security Platform

Mave Raises $34M for Agentic Security Platform

Mave Raises $34M for Agentic Security Platform - Israeli startup Mave Security has raised $34 million to expand its agentic platform, which connects an enterprise's existing security tools into a single system without requiring data to be moved into a central database. The platform queries each tool in its native language, investigates every alert...

AI STARTUP FUNDING
Microsoft AI Outdiagnoses Doctors on Complex Cases

Microsoft AI Outdiagnoses Doctors on Complex Cases

Microsoft AI Outdiagnoses Doctors on Complex Cases - Microsoft's AI diagnostic system MAI-DxO correctly diagnosed complex medical cases 83% of the time in a simulation, compared to 36% for generalist physicians using reference materials, according to a study published in Science. The system coordinates multiple AI agents that question patients, order ...

AI MEDICAL SUPERINTELLIGENCE
Nine Oracle Executives Preview Enterprise AI World

Nine Oracle Executives Preview Enterprise AI World

Nine Oracle Executives Preview Enterprise AI World - Ahead of Oracle AI World 2026 in Las Vegas (Oct. 25–28), nine Oracle executives spoke with theCUBE Research previewing the conference's Data Deep Dive programming. Key themes include moving enterprise AI from experimentation to production, securing agentic AI systems, and database modernization. Exe...

AI ENTERPRISE INSIGHTS
1Password lets AI agents log in without seeing passwords

1Password lets AI agents log in without seeing passwords

1Password lets AI agents log in without seeing passwords - 1Password has launched 1Password for Agents, a sign-in platform that allows personal AI agents from Meta, Anthropic, OpenAI, and SpaceXSI to access accounts without ever receiving the actual password. Instead, a unique digital key identifies the agent, and the account owner must approve each login o...

AI SECURITY LAUNCH
IBM puts governed context at center of AI

IBM puts governed context at center of AI

IBM puts governed context at center of AI - IBM is positioning governed context as a core requirement for production AI agents. Shaun Clowes, general manager of Confluent and Data at IBM, argues that agents need tailored, current information rather than access to all available business data. IBM's streamhouse architecture, built on IBM Conflu...

AI ENTERPRISE GOVERNANCE
Claude Failed to Invent Scorpion Antivenom

Claude Failed to Invent Scorpion Antivenom

Claude Failed to Invent Scorpion Antivenom - An investment research firm partnered with Anthropic to test whether an unreleased version of Claude could develop scorpion antivenom, aiming to demonstrate AI's potential in drug development. The experiment fell short. While Claude produced some promising starting points, experts told The Verge it ...

AI MISINFORMATION CLAIM
OpenAI Brings GPT-6 Models to Google Cloud

OpenAI Brings GPT-6 Models to Google Cloud

OpenAI Brings GPT-6 Models to Google Cloud - OpenAI will make its flagship GPT-6 Astra model and two scaled-down variants, GPT-6.1 Sol and GPT-6 Luna, available through Google Cloud's Model Garden service. Astra achieved a perfect score on the ExploitBench cybersecurity benchmark and includes enterprise-focused safety optimizations. Sol and Lu...

GENERATIVE AI PARTNERSHIP
IBM Beats Q3 Estimates But Mainframe Slump Lingers

IBM Beats Q3 Estimates But Mainframe Slump Lingers

IBM Beats Q3 Estimates But Mainframe Slump Lingers - IBM reported third-quarter adjusted earnings of $2.95 per share, beating Wall Street's $2.86 forecast, with revenue rising 5% to $17.1 billion. Software led growth at 8%, while consulting was flat and infrastructure rose just 4%. AI now accounts for 30% of consulting revenue and 35% of the order bac...

AI ENTERPRISE PERFORMANCE
Discord fixes broken Markdown preview rendering

Discord fixes broken Markdown preview rendering

Discord fixes broken Markdown preview rendering - Discord's latest update improves Markdown file previews, which previously displayed raw symbols instead of formatted text. The fix is particularly relevant for users sharing AI-related content, as large language models and AI agents commonly use Markdown files for instructions and context.

TECH PLATFORM UPDATE
Canva Craft AI plans designs before generating images

Canva Craft AI plans designs before generating images

Canva Craft AI plans designs before generating images - Canva has launched its first proprietary AI image generation model, called Canva Craft, putting it in direct competition with OpenAI, Microsoft, and Google. The model uses a two-stage architecture that plans a design before generating an image, which Canva says delivers greater precision, consistenc...

NEW GENERATIVE AI LAUNCH
Canva launches first in-house image generation model

Canva launches first in-house image generation model

Canva launches first in-house image generation model - Canva has released its first proprietary image generation model, Canva Craft, built by its internal research team with contributions from acquired platform Leonardo.ai. The model works in two stages: interpreting prompts to plan spatial and visual details, then generating images up to 2K resolution ...

NEW GENERATIVE AI LAUNCH
Microsoft AI claims to outdiagnose human doctors

Microsoft AI claims to outdiagnose human doctors

Microsoft AI claims to outdiagnose human doctors - Microsoft says it has developed an AI tool capable of diagnosing medical conditions up to four times more accurately than human doctors. Microsoft AI CEO Mustafa Suleyman told The Verge that the moment AI leaders have long promised for healthcare has finally arrived. The company is positioning the t...

AI HEALTHCARE DIAGNOSIS
Meta fined $25 billion over privacy deception

Meta fined $25 billion over privacy deception

Meta fined $25 billion over privacy deception - A New Mexico jury found Meta guilty of 43 million violations of the state's unfair practices act, stemming from the Cambridge Analytica scandal when the company operated as Facebook. The jury initially awarded $219.5 billion, but Judge Francis Mathew reduced the penalty to $25 billion. The fine grow...

AI REGULATORY UPDATE
Spotify bets on AI agents for music discovery

Spotify bets on AI agents for music discovery

Spotify bets on AI agents for music discovery - Spotify reported 788 million monthly active users and nearly €5 billion in revenue for Q3, but its earnings call focused on a bigger strategic shift. Co-CEO Gustav Söderström argues Spotify is well-positioned for the agentic AI era, since people still want to experience entertainment rather than hav...

AI PLATFORM STRATEGY

Stay ahead of the curve

Get a weekly AI news roundup delivered to your inbox every Monday.