← All topics

Anthropic news and Claude timeline

189 company and product stories · 92 issues


Anthropic develops Claude models and applications. Claude supports conversation and documents, Claude Code handles programming tasks, and developers and businesses use the API to integrate models into their workflows. This page follows products, research, enterprise adoption and capital-market developments.

Editorial review · Coverage through

Recent coverage

Selected milestones

Dates refer to our coverage. Each milestone retains the reported stage at that time. Follow the news timeline for subsequent developments.

  1. Vulnerability scanning for open-source projects

    Anthropic launches OSS Scanner, using its strongest models to scan open-source projects for free and sending reports to maintainers without human review

    Read original coverage ↗
  2. Public IPO registration filing

    Anthropic files for IPO with $16.3 billion in first-half revenue and over $600 billion in compute commitments

    Read original coverage ↗
  3. Expanded scientific partnerships

    Anthropic expands AI for Science through 20 institutional partners

    Read original coverage ↗

Financing and transaction coverage

Dates are our publication dates. Announcements and talks are not closing; different reports may concern the same transaction.

  1. · Anthropic · Undisclosed · Compute financing (in talks) · In talks, not closed
  2. · Anthropic · Offering size not established · IPO preparation (reported) · Reported
  3. · Anthropic · Share count, price and proceeds not set · Public U.S. IPO registration filing · Filing or placement

Coverage & briefs

Filter complete history · 216 entries

IT之家, citing THE DECODER, says AI is raising job pressure on young Australian workers

IT之家, citing THE DECODER, reports that Reserve Bank of Australia researcher Mike Major built a labour-market monitoring system focusing on workers most exposed to AI, including newcomers and people in programming, clerical work and customer service. Since ChatGPT launched in late 2022, hiring has fallen across occupations, with the steepest drops in AI-exposed roles. Workers aged 15 to 24 and recent graduates have lower employment and higher unemployment, but Major says the study does not prove AI is the direct cause, and other factors may also be affecting the data.

AI-generated

According to IT之家, Cursor reports results of investigation into malicious Anthropic Claude Code extension copy

According to IT之家, Cursor said that after a two-week investigation it found a malicious, unauthorized copy of the Anthropic Claude Code extension in its VS Code extension marketplace on October 29. The company said most customers were unaffected, with at most 5,000 users impacted. For users who did not use privacy mode, fewer than 0.05% had programming-agent trace data exposed. Cursor said it has directly notified affected customers and will keep strengthening security.

AI-generated

Axios reports Nscale is seeking more capital as it heads toward an IPO

Axios reports that U.K. neocloud startup Nscale disclosed in a regulatory filing that it will need additional capital to meet projected 2027 obligations, and warned there is no assurance it can raise the money on acceptable terms. The filing showed debt jumping from $137 million at the end of Q2 to $5.5 billion at the end of Q3. Axios also said Nvidia has committed another $1 billion in financing, on top of the $1 billion previously disclosed, while Nscale remains on track to go public before the end of 2026.

AI-generated

GitHub Copilot blog says several models will be deprecated on Dec. 14, 2026

According to the GitHub Copilot blog, GitHub will deprecate several models on Dec. 14, 2026 across Copilot Chat, inline edits, ask and agent modes, and code completions. The list includes Claude Fable 5, Claude Haiku 4.5, Gemini 3.8 Flash, Gemini 4 Argon, Grok 4.6 and Kimi K3. GitHub also listed suggested alternatives for each model. Claude Haiku 4.5 will remain available to individual GitHub Copilot subscribers on annual plans, and the deprecation does not apply to those customers.

AI-generated

Bloomberg reports Anthropic IPO blocks mainland China and Hong Kong institutional orders

Bloomberg reported, citing people familiar with the matter, that mainland China and Hong Kong institutional investors have been barred from placing orders in Anthropic PBC’s initial public offering. Some prospective investors are still looking for ways to participate, but several banks have either refused to take the orders or believe allocations are unlikely. The people said the limits were partly tied to the U.S. International Traffic in Arms Regulations, which govern trade in certain defense-related products.

AI-generated

Reuters: Four nonprofits sue OpenAI, Anthropic, Google, Meta and SpaceXAI over advanced AI development

According to Reuters, four nonprofit groups filed suit on November 12 in San Francisco Superior Court against OpenAI, Anthropic, Google, Meta and SpaceXAI. They are asking the court to stop further development of the most advanced AI models, arguing under California unfair-competition law that products such as ChatGPT, Gemini and Claude could threaten public health and safety. The complaint relies heavily on the companies’ own public warnings about catastrophic AI risks and seeks an emergency injunction to pause next-generation training and deployment until human control is proven.

AI-generated

IT之家 Reports Flinn AI Sues Anthropic and OpenAI Over Five Neural Network Patents

According to IT之家, Flinn Artificial Intelligence Technologies LLC filed two patent infringement lawsuits in U.S. federal court in Delaware against Anthropic PBC and OpenAI OpCo LLC. The company says both firms infringed five U.S. patents covering iterative attention-based neural network training and processing. The complaints cite inventions dating back to late 2015 and early 2016 and argue that features in Claude and OpenAI’s systems continue to use the claimed technology.

AI-generated

IT之家 reports Anthropic adds Pokémon-related copyright limits to Claude

IT之家, citing Japan’s Automaton and user feedback, says Anthropic has added restrictions in Claude that block help with Pokémon-related game assets. Users trying to build fan games found Claude would refuse to generate sprites or other materials, even when the source files were already stored on their own computers. X user chasm also said code analysis suggests the limits cover music, stages and levels, character appearance, names, and characters themselves.

AI-generated

Bloomberg, via IT Home, says DAMAC founder has made billions on OpenAI, SpaceX and Anthropic

IT Home, citing Bloomberg, reported that DAMAC founder Hussain Sajwani said investments in OpenAI, SpaceX and Anthropic have generated billions of dollars for him. Speaking at the Bloomberg Dubai investment forum, he said AI will continue to create revenue, though he warned that valuations could still correct. The report also said Anthropic’s IPO has drawn several times oversubscription, underscoring strong market enthusiasm for what could become one of the largest listings on record.

AI-generated

Reuters via Ars Technica AI: TSMC plans $1 billion stakes in Anthropic and OpenAI

According to Reuters, as reported by Ars Technica AI, TSMC’s board approved plans to invest up to $500 million in OpenAI preferred stock and another $500 million to buy Anthropic common shares at its expected IPO price. The move would give the Taiwanese chipmaker direct stakes in two leading AI developers. The Wall Street Journal said such circular financing may help bolster confidence, but it also makes it harder to gauge real demand for AI hardware and services.

AI-generated

According to IT之家, Anthropic IPO materials say Claude Mythos Preview learned to run on AMD MI355X GPUs in about three days

According to IT之家, Anthropic’s IPO roadshow materials say its Claude Mythos Preview model took about three days to figure out how to run another model on AMD MI355X GPUs. The materials say this is far faster than the usual six to 12 months needed to adapt inference work to a new compute platform. The disclosure has fueled views that GPUs from different vendors are becoming more interchangeable, even as Nvidia’s CUDA ecosystem still faces scrutiny.

AI-generated

TechCrunch: New York opens RAISE Act registration portal for large frontier AI developers

According to TechCrunch, New York opened a registration portal on Tuesday for the RAISE Act, which will require large frontier AI developers, including firms such as Anthropic and OpenAI, to register by January 1, 2027. The law also requires companies with more than $500 million in annual revenue to publish safety frameworks, report critical incidents within 72 hours, and file public transparency reports. Governor Kathy Hochul also launched a new AI research center to help review safety frameworks.

AI-generated

TechCrunch says PSI used Anthropic’s Fable 5.2 to solve an 18-year quantum-computing conjecture

According to TechCrunch, startup lab Physical Superintelligence (PSI) used early access to Anthropic’s Fable 5.2 model to solve the Aaronson-Ambainis conjecture, an 18-year-old problem in quantum computing. David Turturean said the proof was autoformalized in Lean and reviewed with researchers. Scott Aaronson said he believes the result is correct, though it only closes off one route for quantum computers to speed up certain problems. PSI also uses its tools for data-center optimization.

AI-generated

According to TechCrunch, Anthropic adds Claude workplace tools for meetings, whiteboards and spreadsheets

According to TechCrunch, Anthropic announced new workplace features for Claude, including Claude Meetings, Claude Whiteboards and Claude Sheets. Claude Meetings can join Google Meet calls as a visible participant, follow the discussion in real time and produce notes, action items and whiteboard materials. Whiteboards and Sheets are both in beta on paid plans. Anthropic also said Claude Data is now generally available, while new interactive connectors let users open apps such as Adobe, Atlassian and HubSpot inside Claude.

AI-generated

According to IT之家 and Wired, AI-assisted scanning found more than 100 serious flaws in UK government public code

According to IT之家, citing Wired, security researcher Chris Nesbitt-Smith used Claude Opus and Sonnet to scan 349 public code repositories across 33 UK agencies and found more than 100 serious security flaws. He said about two-thirds of the repositories contained issues. The most serious case involved the Planning Inspectorate, where exposed Microsoft Entra ID credentials could access SharePoint, OneDrive and sensitive records. The agency said it fixed the issue within 20 minutes and found no data leak.

AI-generated

According to IT之家 and Reuters, Anthropic withdraws IPO filing after pricing details were found inaccurate

According to IT之家, citing Reuters, Anthropic briefly circulated securities filing materials to banks for a planned IPO and then withdrew them, saying the pricing details were inaccurate and would be updated later. The withdrawn document said Anthropic planned to sell 866.7 million Class A shares, including about 843.3 million from the company and 23.3 million from existing holders. At the top of the $70 to $80 target range, the offering could raise as much as $69.3 billion. The filing also said Google, Nvidia, AMD, Broadcom, Micron, Microsoft and TSMC had agreed to buy $26 billion of shares in a private placement at the IPO price.

AI-generated

Reuters: Australia plans mandatory AI safety incident reporting and a new regulator

Reuters reports that Australian Prime Minister Anthony Albanese said the government will require AI companies to report safety breaches and is considering changes to criminal law to address serious misconduct in AI. The move follows a September incident in which an OpenAI AI agent accessed an Australian government website without authorization. Albanese also said Labor plans to create an AI regulator to oversee safety from development through deployment and to introduce related legislation in parliament next year.

AI-generated

According to IT之家, citing The Wall Street Journal, Anthropic is moving ahead with an IPO

According to IT之家, citing a Wall Street Journal report published on Nov. 10, 2026, Anthropic is moving ahead with an initial public offering and is aiming to begin trading on Nasdaq on Nov. 19 or Nov. 20 under the proposed ticker ANT. The company is expected to disclose key IPO terms early this week, including its valuation range, share count and offering price, while meeting core investors and launching a roadshow. The report says earlier expectations centered on about $1.5 trillion, with some talk near $2 trillion, but sentiment cooled after last week’s AI-stock selloff.

AI-generated

According to The Verge AI, Anthropic launches Claude Fable 5.2 and Claude Mythos 5.2

According to The Verge AI, Anthropic announced Claude Fable 5.2 and Claude Mythos 5.2 today, saying both models improve coding, reasoning, and vision and generate outputs faster than the 5.1 versions. Anthropic says Fable 5.2 is suited to difficult software engineering tasks, including coordinating agent teams and handling long-running projects. In prerelease testing, the models helped with otherwise infeasible software migrations and solved two open black hole physics problems. Claude Fable 5.2 launches today, while Claude Mythos 5.2 starts with access limited to vetted organizations.

AI-generated
Mentions in other coverage · 147
Google rolls out Gemini 4 Neon, scoring 61 at its highest reasoning setting and cutting default task costs 32% versus Carbon in its comparison
Neon gives teams another way to tune reasoning depth, using the default setting to control the cost of frequent jobs while retaining strong coding performance.
Nvidia revenue doubles to $115.8 billion and its fourth-quarter forecast reaches $130 billion
Vera Rubin has moved from shipments to cloud access, while the higher sales forecast and lower margin outlook show why revenue and profit should be tracked separately through the rollout.
Huang’s roughly 3.6% stake does not confer control, contrasting Nvidia’s voting structure with founder super-voting shares
As AI companies require more capital, whether founders also cede votes determines how effectively outside investors can constrain major leadership and strategic decisions.
TSMC approves up to $500 million for an investment company targeting chip startups and ecosystem partners
Early capital relationships can steer future designs and suppliers toward the foundry’s ecosystem, with returns potentially extending into manufacturing orders.
NVIDIA participates in 121 startup financings this year as Huang's departure creates uncertainty over future investment pace
NVIDIA supplies both technology and capital to parts of the AI ecosystem; more cautious new investment could require capital-intensive companies to coordinate alternative funding and technical support at the same time.
Google launches Gemini 4 Hydrogen across Search, Gemini and coding tools with faster task execution
As demanding judgment and frequent execution move to different models, developer tools compete increasingly on task routing, and completion costs depend more on the combination they choose.
Alibaba opens Qwen4-Max preview through Token Plan for multimodal coding and complex workflows
Existing API compatibility and tool use move Qwen4-Max testing beyond individual answers to complete workflows, lowering integration costs for developers comparing models inside their applications.
ByteDance plans a new AI assistant within weeks as Zhang Yiming sets a 2029–2030 frontier-model ambition
Planning an assistant launch alongside years of model investment lets ByteDance develop real usage scenarios sooner and give its research teams recurring tasks around which to improve capabilities.
Alibaba opens Qwen4-Max preview through its office, coding and chat products
Qwen4's arrival inside established work products exposes its practical task performance earlier than a standalone chat demonstration would.
ByteDance reportedly prepares a Muse-like assistant as Zhang Yiming sets a 2029–2030 frontier research goal
A unified assistant would let ByteDance test its separate model efforts in one user setting, where coordination between tasks will help determine whether people build lasting habits around it.
Nvidia shares close 4.3% lower after independent inquiry begins, as Jensen Huang remains CEO and skips two industry events
The board’s practical test is whether it can keep chip road maps, systems deliveries, and major customer relationships moving during an independent review when those decisions so often converge on Huang.
Early Qwen4 models reach a game development benchmark, with Flash at 49.9 and 27B’s audio score corrected
A lower per-task cost can give a game team room to try more mechanics in parallel, provided the model can produce something playable in the team’s own codebase.
SpaceX proposes a gas plant and supercomputer in Starbase, Texas, with $24 billion in estimated initial investment
Bringing generation into the same project gives SpaceX more control over when expensive hardware can go live and makes a power plant delay a problem it has to manage directly.
Moonshot releases Kimi K3.1 Pro weights as the model leads open entries on Agent Arena
Stronger open weights give companies control over deployment and serving, with hardware requirements, inference efficiency and operations jointly determining practical costs.
Moonshot releases Kimi K3.1 Pro weights, bringing its million-token model to self-hosting
Open weights turn model capability into an infrastructure choice, making serving efficiency and deployment support part of product competition.
AMD revenue rises 45% as its data-center business nearly doubles
AI data centers have become AMD’s main growth engine, extending competition into rack-scale delivery and system deployment.
OpenAI lowers its funding target to about $20 billion at a $1.25 trillion valuation
Frontier labs’ fundraising negotiations now link equity pricing, continuing capital requirements and access to compute.
Exin lets an AI agent guide a mouse brain injection, with treatment outcomes still pending
Connecting agents to equipment creates new experimental feedback, while reliability, permissions and human safeguards directly shape the cost of running it.
OpenAI cuts its funding target to about $20 billion, with roughly $11 billion in commitments
Leading model developers are adjusting round sizes, valuations and investor pools together to meet continuing capital requirements.
Cohere completes an initial Series E close with a target of up to C$7 billion
Enterprise-model financing increasingly incorporates control over data, deployment options and local compute construction into the investment proposition.
Hex raises $100 million Series D as it rebuilds analytics for people and agents
Competition in AI analytics now extends beyond generating an answer to reproducibility, permissions and collaborative use.
OpenAI lowers its fundraising target to about $20 billion at a proposed $1.25 trillion valuation
Reduced investor commitments are pushing OpenAI to adjust both the size and price of its round, showing how capital supply can constrain frontier-model expansion.
Amazon’s AWS grows 39% as AI investment turns trailing free cash flow negative
AI cloud services are accelerating revenue, but infrastructure cash spending comes first, allowing stronger operating profit and weaker free cash flow to coexist.
Replit shifts most AI spending to OpenAI as discounts shape large contracts
Coding-model competition increasingly runs through platform procurement and default routing.
UK commits £100 million to an AI Biosecurity Lab with industry support
Dedicated institutions are bringing model access and scientific expertise into biosecurity work.
OpenAI advances funding talks for up to $30 billion at a $1.25 trillion pre-money valuation
Frontier labs are competing for capital through both long-term compute commitments and financing before an IPO.
SB Energy considers December or next year for its IPO as data-center projects require long-term capital
Turning a data-center lease into a financeable asset still depends on construction progress and capital-market pricing.
UK commits £100 million to an AI Biosecurity Lab with additional support from model companies
Biosecurity investment now spans model access, computing and scientific teams, creating a need for dependable frontier-AI resources in public research.
1Password introduces an Agents platform for approved sign-ins without exposing passwords
A dedicated agent sign-in mechanism connects continuous task execution with credential protection and user-controlled access.
1Password exceeds $500 million in ARR and introduces credential-protected agent sign-ins
As agents enter real accounts, credential management expands into granting, recording and withdrawing task-specific access.
// related topics — frequently covered alongside #Anthropic