Anthropic nears $1 trillion as governance frameworks multiply
Anthropic's $65B funding round vaults it past OpenAI; new models, governance frameworks, and AI security investment signal industrial maturity.
By BINA Editorial
Today's AI landscape shows a sector in rapid commercial and institutional maturity: a record near-trillion-dollar valuation, a new flagship model, and multiple organisations publishing safety and governance frameworks in the same week.
Anthropic nears $1 trillion with record $65 billion funding round
Anthropic has closed a $65 billion Series H funding round, pushing its post-money valuation to $965 billion and surpassing OpenAI's $852 billion figure from March. The round was led by Altimeter Capital, Dragoneer, Greenoaks, and Sequoia Capital, with existing hyperscaler commitments from Amazon included. The company's run-rate revenue has reportedly climbed to more than $47 billion — up from $14 billion in February — driven by enterprise adoption of its Claude models.
Anthropic releases Claude Opus 4.8 with faster mode and dynamic workflows
Anthropic launched Claude Opus 4.8 on May 28, now available on claude.ai, Amazon Web Services, and Microsoft Foundry. The update adds user controls over model effort, a "dynamic workflows" feature within Claude Code for large-scale tasks, and a faster operating mode running at 2.5 times the speed and one-third the cost of previous configurations. Standard pricing remains unchanged: $5 per million input tokens and $25 per million output tokens.
OpenAI publishes Frontier Governance Framework for regulators
OpenAI released its Frontier Governance Framework on May 28, a public document mapping its safety and security protocols to emerging legal requirements — specifically California's Transparency in Frontier AI Act and the EU AI Act's Code of Practice for General Purpose AI. The framework covers risk assessment across domains including cyber-offence capabilities, CBRN risks, harmful manipulation, and potential loss of control, and outlines how OpenAI will engage with external auditors and update its practices as models and regulations evolve.
EDF and Mistral AI partner to bring AI to French nuclear engineering
France's EDF and Paris-based AI company Mistral AI signed a five-year partnership on May 28 to develop AI tools for EDF's nuclear engineering, maintenance, and construction activities — including future EPR2 reactors. The collaboration will focus on conversational agents capable of querying the technical memory of the French nuclear fleet and construction sites. All data will remain EDF's property and be hosted on sovereign cloud infrastructure or EDF's own data centres; the AI tools will not be used in nuclear plant control systems.
IBM and Red Hat commit $5 billion to secure open-source AI infrastructure
IBM and Red Hat announced Project Lightwell on May 28, a $5 billion initiative to harden open-source software against AI-accelerated vulnerability discovery and exploitation. A global team of over 20,000 engineers will use frontier AI tools to identify, validate, and deploy patches through a trusted enterprise clearinghouse. Bank of America, JPMorgan Chase, and Visa are among the first institutions joining the service.
US Federal Reserve official cautions against counting on AI productivity to fight inflation
St. Louis Federal Reserve President Alberto Musalem warned on May 28, speaking at a conference in Reykjavik, that monetary policymakers should not rely on an AI-driven productivity surge to bring down US inflation, which currently stands at 3.8% — well above the Fed's 2% target. Musalem's caution reflects growing concern in central banking circles that AI's macroeconomic effects remain uncertain in both timing and magnitude, and that premature optimism about future productivity gains risks delayed action on present price pressures.