Substrate
ai

Anthropic in Talks to Raise Tens of Billions at $950 Billion Valuation

Anthropic is in talks to raise tens of billions of dollars in a funding round that would value the company at $950 billion, surpassing OpenAI's $854 billion valuation from March. The AI developer has quadrupled its market share among business customers since May 2025 and released nearly as many models in 2026 as all of 2025.

Techcrunch
DI
SK
GB News
Wired
The Guardian
6 sources·May 13, 8:19 PM(1 hr ago)·3m read
|
Anthropic in Talks to Raise Tens of Billions at $950 Billion Valuationbbc.co.uk
Audio version
Tap play to generate a narrated version.

Anthropic is looking to raise tens of billions of dollars in a funding round that would value the company at $950 billion. OpenAI was valued at $854 billion in its March round. The AI developer recently outpaced OpenAI among business customers and quadrupled its market share since May 2025.

Cat Wu joined Anthropic in August 2024 as head of product for Claude Code and Cowork. She has played a central role in shifting Claude from an informational chatbot to a coding tool. Wu sat down for an interview at the second annual Code with Claude conference, which took place in San Francisco in the week prior to May 13 2026.

Anthropic released at least six models in 2025 and has already released almost as many in 2026. Wu said the company hopes this pace of development continues because the models are still improving at a steady pace. "For us, we just need to stay at this frontier," she stated, adding that focusing on competitors would leave the team perpetually behind.

The company launched the Glasswing initiative in April 2026. Glasswing gave a small consortium of partner organizations including Amazon, Apple, CrowdStrike, and Microsoft access to Anthropic's new cybersecurity model Mythos. Mythos is designed to scan codebases for software vulnerabilities and is not being given a general public release because Anthropic fears it could be weaponized by bad actors.

Wu described managing AI agents as similar to managing people, requiring domain expertise to debug mistakes and clarify under-specified requests. She said the long-term goal is not simply to cut team sizes but to eliminate tedious work so employees can focus on creative building.

In the next six months, she expects proactivity to be the next big advance, with Claude anticipating needs and setting up automations without explicit instructions.

Separately, WhatsApp is launching an AI chat function known as Incognito Chat. Incognito Chat is built to allow users to converse privately with Meta AI such that Meta itself cannot access the questions or answers. With Incognito Chat, WhatsApp will only be able to see that an account used the feature.

Incognito chats are ephemeral by default and will disappear once the conversation is over. The feature is based on WhatsApp's Private Processing scheme which debuted a year ago and already underlies existing AI features including message summarization and composition tools. Incognito Chat is currently text-only with image processing and voice recognition support in development.

Meta is adding Incognito Chat as an option in the Meta AI app. The company announced “Side Chat with Meta AI” which allows users to DM with Meta AI about a text chat they are having with another person or group. Incognito Chat includes a feature turned on by default that allows its AI model to search the web in an anonymized way for up-to-date information.

Mark Zuckerberg wrote that Incognito Chat handles all AI inference in a Trusted Execution Environment that ensures messages are not accessible to Meta and that conversations on the phone disappear when the session is exited. WhatsApp has more than 3 billion users around the world.

Will Cathcart, head of WhatsApp, said the challenge was building data-center AI with the security properties of a device that fits in a pocket.

More than 25 states and multiple countries have enacted laws requiring mixed-audience websites to verify users’ ages to prevent access by children. Some age verification laws target adult content providers explicitly while others apply to a wide array of websites including Google search, Coursera, and the New York Times.

Requirement to upload documents for the purpose of ‘age verification’ is coming to platforms.

In unrelated comments, Jake Gyllenhaal said he enjoys working with Guy Ritchie as original movies become more of an impossibility. " Lord Peter Ricketts said Starmer's leadership challenge is a real blow to Britain's standing on the global stage.

Key Facts

Anthropic funding round at $950 billion valuation
The company is seeking tens of billions of dollars, exceeding OpenAI's $854 billion March valuation after quadrupling business market share since May 2025
Incognito Chat launch on WhatsApp
Feature uses Trusted Execution Environment and Private Processing so Meta cannot access conversations; chats are ephemeral by default with web search enabled
Glasswing initiative and Mythos model
Limited access granted in April 2026 to Amazon, Apple, CrowdStrike and Microsoft for vulnerability scanning tool withheld from public due to weaponization risks

Story Timeline

5 events
  1. 2024-08

    Cat Wu joined Anthropic as head of product for Claude Code and Cowork

    1 sourceTechCrunch
  2. 2025-05

    Anthropic began quadrupling its market share among business customers

    1 sourceTechCrunch
  3. 2025

    Anthropic released at least six models

    1 sourceCat Wu
  4. 2026-04

    Anthropic launched the Glasswing initiative giving partners access to Mythos model

    1 sourceTechCrunch
  5. 2026-05

    Second annual Code with Claude conference held in San Francisco; WhatsApp announced Incognito Chat

    2 sourcesTechCrunch · Meta

Potential Impact

  1. 01

    Age verification document upload requirements will affect mixed-audience sites from search engines to news outlets

  2. 02

    Privacy-focused Incognito Chat may accelerate mainstream AI adoption among WhatsApp's 3 billion users

  3. 03

    Accelerated model releases and proactive agent features could reduce tedious workplace tasks across industries

Transparency Panel

Sources cross-referenced6
Framing risk55/100 (moderate)
Confidence score97%
Synthesized bySubstrate AI
Word count639 words
PublishedMay 13, 2026, 8:19 PM
Bias signals removed2 across 2 outlets
Signal Breakdown
Loaded 2

Related Stories

Robot Navigation Strategy Based on Honeybee Learning Flights Proposed in Nature Paperinterestingengineering.com
ai1 hr ago

Robot Navigation Strategy Based on Honeybee Learning Flights Proposed in Nature Paper

A research paper published in Nature proposes Bee-Nav, which trains a tiny neural network during short robotic learning flights to enable efficient visual homing. Real-world tests with a small drone achieved high success rates over distances up to 600 meters using neural networks…

NA
Inside Climate News
Hoover Institution
3 sources
Anthropic in Talks to Raise Funding at $950 Billion Valuationnewser.com
ai1 hr ago

Anthropic in Talks to Raise Funding at $950 Billion Valuation

Anthropic is in discussions to raise funding that would value the artificial intelligence start-up at $950 billion. The company was previously valued at $380 billion. It recently released a new AI model called Mythos and is separately in a dispute with the Pentagon.

The New York Times
economictimes.indiatimes.com
zerohedge.com
newser.com
4 sources
Trump and Xi to Discuss AI, Chip Exports and Supply ChainsRest of World
ai1 hr ago

Trump and Xi to Discuss AI, Chip Exports and Supply Chains

U.S. President Donald Trump and Chinese President Xi Jinping are scheduled to meet in Beijing this week. The summit is expected to cover artificial intelligence competition, advanced chip sales, supply chain security, electric vehicle trade and rare earth minerals. Discussions co…

SE
SQ
Rest of World
3 sources