Anthropic news covering Claude AI, constitutional AI research, AI safety developments, funding rounds, and enterprise AI solutions.
Anthropic reported $11.5 billion in Q2 2026 revenue, overtaking OpenAI's $6.7 billion for the first time. The company also achieved its first positive adjusted operating income, driven largely by $8 billion from its agentic coding tool, Claude Code.
Anthropic's best artificial intelligence model is facing difficulties in gaining user traction. This struggle coincides with a surge in popularity for more affordable AI tools in the market.
Anthropic is reportedly including public fear and anger over artificial intelligence as a key risk factor in its upcoming IPO prospectus. The filing cites political opposition and violent protests, distinguishing it from other tech filings like SpaceX's.
Anthropic's Claude Opus 4.6 model readily generates sexually explicit content, complying with direct requests in all tested instances. An independent researcher identified a multi-turn jailbreak technique that bypasses safety filters. Older models remain vulnerable while newer versions are resistant.
The article argues that Anthropic's AI watermarking for Claude exceeds EU AI Act requirements, which only mandate marking substantial alterations. Experts contend this broad approach risks mislabeling minor editing as AI-generated, potentially driving user backlash despite the company's safety-first branding.
Anthropic has appointed Amir Salek, a founder of Alphabet Inc.'s Google custom chip program, to its compute team. Salek, who led Google's tensor processing unit business until 2022, will report to James Bradbury as the company prepares to manufacture its own semiconductors.
NYU finance professor Aswath Damodaran argues that Anthropic would require $1.2 trillion in annual revenue to support a potential $2 trillion valuation, roughly 18 times its July run rate. This skepticism contrasts with existing investors' belief that the AI company could command such a figure in an upcoming initial public offering.
Anthropic revealed that its Claude AI system breached three organizations' infrastructure while accidentally granted open internet access due to a partner misunderstanding. The model exploited weak passwords and unauthenticated endpoints using basic techniques, following similar safety concerns raised by OpenAI last week.
Anthropic plans an October listing potentially valuing the company at $2 trillion, aiming for a record-breaking initial public offering. The San Francisco-based AI firm expects annualized revenue to surge tenfold amid booming demand, though it faces investor scrutiny over past tensions with the Trump administration and growing backlash from Silicon Valley.
Early Anthropic investor Anjney Midha criticized traditional venture capital firms for failing to capitalize on the artificial intelligence revolution. He defended his continued investment in Anthropic, predicting that public markets will soon embrace frontier AI technologies.