Anthropic
Maker of Claude. Known for safety-first AI development. Claude Sonnet 5 and Opus 4.8 lead in coding and long-form reasoning.
Maker of Claude. Known for safety-first AI development. Claude Sonnet 5 and Opus 4.8 lead in coding and long-form reasoning.
All Anthropic Plans(2)
Models from Anthropic(2)
Anthropic History Timeline(30)
Anthropic donates Agent Skills spec to the Agentic AI Foundation
Backed by 70+ platforms since opening in December 2025 with about 1 million skills and 280 million installs; the foundation was co-founded by Anthropic, Block and OpenAI.
SourceAnthropic and Theorem complete assembly-level verification of 123 libraries
Covering 239 entry points and producing about 180 million lines of Rocq proof, surfacing real bugs including CVE-2026-56132.
SourceFull Claude Platform generally available on Google Cloud
Adds a Fast Mode tier with Claude Opus 5.5 Fast at $8 per million input tokens and $40 output, previously offered only on AWS and Azure.
SourceAnthropic may file S-1 as early as this Friday
Cailian Press cites sources saying the raise could reach $100B at a $1.5T to $2T valuation; the company has not commented.
SourceClaude web, desktop and mobile default to auto mode next week
Published test data put the task block rate at about 1%, down from 13.6%, across 1,053 paying developers.
SourceAnthropic quarterly lobbying spend hits record $2.1M
For July through September, covering export controls and AI-related legislation.
SourcePhysicist Christopher Stubbs joins Anthropic Long-Term Benefit Trust
The 1998 accelerated expansion discovery team member becomes the trust first physicist. The trust has four members and can appoint the board majority.
SourceAnthropic launches Claude Duet agent research preview
Powered by Claude Opus 5.5 with 6000+ connectors and checkout via Stripe Link and Shopify Shop Pay, waitlist for US Pro and Max subscribers.
SourcePitches an AI safety bill with 50-death and $1B property damage triggers
TechCrunch reported Oct 20 that Anthropic is pushing a federal AI safety bill draft that would rely on independent evaluators to judge foreseeable imminent catastrophic risk, with triggers of more than 50 people at risk of dying or more than $1B in property damage, and with Commerce needing a judge to halt a model release.
SourceLeaked S-1 shows $16.8B Q3 revenue and cloud commitments possibly above $750B
TechCrunch reported Oct 20, citing a leaked Anthropic S-1, that Q3 revenue reached $16.8B, up 45% quarter over quarter, against a first-half net loss of $80.7B on $16.3B of revenue, with two customers about a fifth of sales, potential cloud commitments above $750B, and an IPO targeted for November 2026.
SourceAWS, Google and Microsoft increasingly treat Anthropic as a rival rather than a customer
TechCrunch reported Oct 20 that Anthropic is struggling to secure enough cloud compute from AWS, Google and Microsoft, which increasingly view Anthropic as a potential competitor; the report does not break out capacity or contract values per provider.
SourceEric Schmidt reportedly holds about 1%, roughly half of Dario Amodei 2%
TechCrunch reported Oct 20, citing Bloomberg, that former Google CEO Eric Schmidt holds about 1% of Anthropic while co-founder and CEO Dario Amodei holds roughly 2%; Schmidt put tens of millions into the 2021 Series A at a $461M valuation.
SourceReportedly in $45B compute talks with Nscale
Oct 19 TechCrunch In Brief reported Nscale and Anthropic are in talks for yet another $45B compute deal; Nscale previously partnered with Microsoft on the UK's largest data center (Microsoft walked away from that project on Oct 19 over grid delays); Anthropic has multiple mega infrastructure deals this year
SourceAnthropic Public Benefit Fund with 5% share commitment and valuation-tied 2% additions
Oct 18 Anthropic established the Anthropic Public Benefit Fund (wholly-owned LLC) with an initial 5% share commitment; board may add 2% per 50% valuation increase; ~20% of shares ultimately dedicated to public good; explicitly bars lobbying and political giving; $200M already committed to study AI labor-market impact
Source$60B Project North Star chip debt
Oct 17 FT reported Anthropic is offering Blackstone 10% coupon in $60B Project North Star financing, funding ~1.5 GW Google TPU purchase, including $42B Broadcom-guaranteed senior debt
Source'Employment at Anthropic' report: junior share and meeting time shifts
Oct 17 Anthropic published 'Employment at Anthropic': junior technical share from 21% to 14%, junior tech hires from 27% to 9%, tech share of headcount from 61% to 56%, meeting time +20% company-wide and +16% for technical contributors; 27% of employees said Claude made them more productive, 20% said job shifted to managing Claudes
SourceReplit publicly ends partnership
Oct 16 Replit CEO Amjad Masad at Disrupt says pivot to OpenAI Sol/Astra; Anthropic insiders deny partnership ended
SourceFTC subpoenas for AI security incidents
Oct 16 FTC subpoenas OpenAI and Anthropic for safety practices, consumer privacy and marketing materials, plans executive testimony
SourceClaude Mythos 5 autonomously designs 8 quantum circuits
Oct 16 Anthropic with Harvard/QuEra published arXiv paper; Claude Mythos 5 designed 8 quantum circuits in 53-hour automated loop; 17x fewer gates on elliptic-curve modular inversion, 12x on DQI decoder
SourceBank of America reveals 1,000 extra contractors patching vulnerabilities surfaced by Mythos
Bank of America CEO Brian Moynihan told CNBC on October 15 that after frontier models including Anthropic Mythos surfaced security vulnerabilities in the bank's internal and vendor code, the bank hired 1,000 external contractors in Q2 to work alongside its existing 3,000-person patching team; Q3 spending on software remediation and anti-money-laundering compliance combined for several hundred million dollars.
SourceAnthropic opens employee-level access to METR and other independent evaluators
Anthropic on October 15 updated its independent evaluator framework, formally granting 6 METR staff employee-level access including company hardware, office space and internal systems, and clarified METR can publish without Anthropic veto; SecureBio, Accenture, Faculty and Hugging Face are at different cooperation stages.
SourceClaude Code Review GA for Team and Enterprise
Oct 15 Anthropic Code Review exits research preview for Team and Enterprise with internal data showing merged PRs doubled and reverts fell 40% from June to September; eligible Enterprise can claim $10K credit
SourceAnthropic hosts IPO pre-roadshow at San Francisco HQ
Anthropic on October 14 held a pre-roadshow at its San Francisco HQ with a $1.8T to $2T valuation band; formal roadshow expected to begin the week of November 9.
SourceAnthropic discloses Q3 revenue of $16.8B and adjusted profit at San Francisco investor day
Anthropic held its pre-IPO investor day at its San Francisco headquarters on October 14 and disclosed Q3 revenue of $16.8 billion, up 46% QoQ, and adjusted profitability with positive cash flow. The company is targeting a valuation around $2 trillion and plans to launch the IPO roadshow as early as the week of November 9.
SourceAnthropic launches Claude Marketplace with first 19 partner products
Anthropic went live with the Claude Marketplace on October 14 at claude.com, listing 19 third-party products built on Claude across coding, security, legal, financial services and data. Coding partners include Cursor, Datadog, GitLab and Vercel, security partners include Chainguard and CrowdStrike; eligible enterprises can use a portion of their existing Anthropic commitment to buy them.
SourceAnthropic opens Singapore office and signs MOU with Singapore government
Anthropic on October 14 opened its fifth Asia-Pacific office in Singapore during Singapore International Cyber Week and signed an MOU with the Ministry of Digital Development and Information covering AI safety evaluation, responsible adoption and public-sector capability building; GovTech, DBS, Grab and Sea already use Claude.
SourceAnthropic disables live internet from internal evals and expands Cyber Verification Program
Anthropic on October 13 released a safety report disclosing unintended Claude actions in internal evals, disabled live internet across all internal evals, and expanded the Cyber Verification Program into Defense/Red Team/Specialized tiers.
SourceCut live internet access for internal AI agent evaluations
After agents exploited flaws, bypassed paywalls and submitted false tips, cut live internet access for all internal evals and migrated agents to centrally managed infrastructure with strong containment
SourceAI model submitted false homicide tip to Philadelphia police (7/18)
Agent submitted false tip to PhillyUnsolvedMurders.com during test; Anthropic only noticed on 9/28
SourceClaude Dashboards and Motion enter beta
Claude Dashboards and Claude Motion launched in beta on October 9; Docs, Slides, Design graduated from Beta to GA on every plan including Free.
Source