Broadcom’s $42B Loan to Anthropic: The 2026 Compute Financing Play
The Broadcom Anthropic $42 billion deal lets Claude’s maker lease custom AI chips on credit. Here’s what it means for compute costs and AI pricing in 2026.
News
Breaking AI news, model releases, funding rounds, regulatory action, and major moves at the leading AI companies.
Cohere’s $7B Valuation & AMD Pact 2026: Sovereign AI Bet
Cohere sovereign AI funding hits $7B on government deals, plus an AMD silicon pact — what a non-Nvidia, non-hyperscaler enterprise LLM stack means for you.
Mistral AI’s €1.7B ASML Round 2026: Europe’s Sovereign Bet
Inside the Mistral AI ASML funding round: €1.7B, an €11.7B valuation, and what Europe’s sovereign AI bet means for business owners choosing a vendor stack.
Emerald AI’s $100M Round 2026: Nvidia Bets on Grid Software
Nvidia’s venture arm anchored Emerald AI’s $100M round in 2026, betting that grid-flexibility software, not more chips, is what unlocks AI data center growth
GPT-6 Astra Debuts 2026: Migrate Your GPT-5 Apps Right
GPT-6 Astra is here, and launch week is the worst time to repin blindly. See what changed in the API, how pricing compares, and how to migrate GPT-5 apps safely
Claude Fable 5.1 Ships 2026: 75% Cheaper Cache Reads
Claude Fable 5.1 cuts cache reads 75%, and with Anthropic users reportedly overpaying 80% on tokens, prompt structure now decides your bill
ChatGPT Ads Hit $1B in 2026: What Advertisers Get Now
OpenAI’s ad business hit a $1B run rate and self-serve ChatGPT ads are now live in Europe. Here’s what you can buy, what it costs, and why the auction is still cheap
AWS + Nvidia Add 2M GPUs in 2026: What Builders Get
AWS and Nvidia are adding 2 million GPUs in 2026 on Vera Rubin silicon — here’s what changes for builders: instance types, regions, and reservations that actually land capacity
Cluely Rival Sett Raises $27M for AI Bot Builders 2026
Sett just raised $27M and opened its AI creative agent to all advertisers — here’s what automated ad variant testing means for your paid traffic economics
Gemini 3.5 Transcribe Ships 2026: Speaker-Aware Notes
Google’s Gemini 3.5 Transcribe merges speech-to-text, speaker labels, and word-level timestamps into one call — here’s what shipped and how to wire it up
Cursor’s Anysphere Buys Graphite 2026: What Changes for Devs
Cursor maker Anysphere is acquiring Graphite, the stacked-PR and AI code review startup. Here’s what the deal changes for dev teams, pricing, and review tooling
Higgsfield Popcorn Ships 2026: 8-Frame AI Story Test
Higgsfield Popcorn generates 8 linked frames from one character reference, so your hero survives the cut. Here’s what shipped, what it costs, and where it still breaks
Unitree’s $5,900 G1 Humanoid Ships 2026: What It Can Do
Unitree’s G1 humanoid robot starts near $16,000 — with entry configs pushing toward $6,000. Here’s what it can actually do, and why the price shift matters
Cluely’s Rival Halo Ships Meeting Copilot 2026: Real Test
Halo AI meeting assistant listens at the OS audio and screen layer, so it never joins your Zoom or Teams call as a bot. Here’s how the Glass client actually tests out
Firecrawl v3 Launches 2026: Agentic Web Scraping Tested
Firecrawl v3 brings agentic crawling that handles logins, infinite scroll, and SPAs with no selectors, plus schema-typed JSON extraction. Here’s how it tested
DeepSeek R2 Ships in 2026: China’s Cheap Reasoning Bet
DeepSeek R2 is an open-weight reasoning model matching frontier US systems on math, coding, and agentic tasks at a fraction of the API cost. Run it now
NVIDIA Vera Rubin Ships 2026: Blackwell’s Successor Arrives
NVIDIA Vera Rubin ships late 2026 with HBM4 memory, a Rubin CPX inference chip, and the Vera CPU. See what’s real, what’s hype, and how to plan 2027 compute
GPT-Live Ships in 2026: OpenAI’s Full-Duplex Voice AI
OpenAI shipped GPT-Live on July 9, 2026 — a full-duplex voice AI that listens, speaks, and reasons at once. Here’s what changes for voice products and support teams
Gemini Enterprise Launches 2026: Google’s AI Agent Platform
Gemini Enterprise launches at Google Cloud Next ’26: Google’s governance-first AI agent platform to build, orchestrate, and supervise agents across every team, tool, and workflow you run
Siri’s Gemini Brain Ships in 2026: Apple’s AI Reboot
The Apple Gemini Siri 2026 deal puts a custom Google Gemini model inside a rebuilt Siri on Private Cloud Compute. Here’s what changes for every iPhone user
Claude Opus 4.8 Ships 1M Context: Anthropic’s 2026 Coder
Claude Opus 4.8 ships a 1M-token context window on Anthropic’s most capable coding model, plus a new Fast mode for Opus-grade output at higher speed
Claude Sonnet 5 Launches: Near-Opus Coding at $2 Per Million
Anthropic’s Claude Sonnet 5 hits $2 per million input tokens with coding and tool-use performance close to Opus 4.8 — here’s what the new price-per-capability math means
AI Agent Software Spending to Hit $206 Billion in 2026, Gartner Forecasts
Gartner projects AI agent software spending will reach $206.5 billion in 2026, up 139% from 2025. Here’s what’s driving the surge and what it means for you.
Meta Llama 3.1 Unveils with 400B Model, New Benchmarks
Meta Llama 3.1 is here! Explore the new 400B model, pushing open-source LLM boundaries & challenging proprietary AI. Get state-of-the-art capabilities.
Meta Muse Spark Debuts From Wang Superintelligence Labs
Meta Muse Spark, the first flagship model from Alexandr Wang Superintelligence Labs, claims frontier-class multimodal and agentic performance at low compute cost.
Google Pics Launches: AI Design Studio Takes On Canva
Google Pics ships at I/O 2026: AI design app powered by Nano Banana 2, comment-driven editing, native Workspace integration, AI Ultra-only at $200/mo.
Karpathy Lands at Anthropic to Lead AI-Assisted Pre-Training
Andrej Karpathy announced May 19 he joins Anthropic to lead AI-assisted pre-training research using Claude — biggest AI talent move of 2026.
‘Dreaming’ Lands: Self-Improving Memory in Claude Managed Agents
Anthropic ships Dreaming May 6: self-improving memory consolidation for Claude Managed Agents with human-approval diff workflow and outcomes tracking.
Grok Build Beta: xAI Launches 8-Parallel-Agent Coding CLI
xAI launches Grok Build May 14: 8 parallel coding agents, plan-search-build workflow, grok-code-fast-1 model at $0.20/M tokens, local-first execution.
Google Ships Gemini Spark: 24/7 Browser Agent at $100/Month
Google ships Gemini Spark at I/O 2026: 24/7 cloud agent on Gemini 3.5, AI Ultra cut to $200, Workspace+Canva+Instacart integration, Chrome agent next.
Grok 4.3 Lands on xAI API With Voice, Imagine, 1M Context
Grok 4.3 xAI API release: 1M context, native video, top agentic tool calling, aggressive pricing, plus Voice and Imagine companion APIs. Frontier-tier developer stack.
Claude Platform Lands on AWS With IAM and Native Billing
Anthropic launches Claude Platform on AWS: full Anthropic API surface with AWS IAM auth and AWS Marketplace billing — distinct from Bedrock. Enterprise impact.
OpenAI Closes Historic $122B Round at $852B Valuation
OpenAI closes $122B funding round at $852B valuation — largest private raise in tech history. Retail IPO carve-out signaled. Anthropic, Microsoft, market impact.
Decart Lands $300M Series B to Break Nvidia CUDA Lock-In
Decart raised $300M Series B at $4B valuation on May 18, 2026. Cross-chip AI portability product (DOS) backed by Nvidia, Radical, Sequoia, Karpathy.
Martha Stewart Launches Hint, an AI Home Management Startup
Martha Stewart launched Hint on May 13, 2026: an AI home management platform with 0M seed funding. Proactive maintenance, insurance, repairs.
OpenAI Launches ChatGPT Personal Finance With Bank Connections
OpenAI launched ChatGPT personal finance on May 15, 2026: Plaid integration, read-only bank access, transaction analysis, budget projections, tax prep.
Meta Launches WhatsApp Incognito Chat for Private AI
Meta launched Incognito Chat for WhatsApp on May 13, 2026 — confidential-compute AI conversations that disappear by default. How it works and how to use it.
Anthropic Ships Claude for Legal With 12 New AI Plugins
Claude for Legal expanded May 12 — 12 role-specific plugins, Microsoft 365 embed, DocuSign/Westlaw/Harvey integrations, broad legal workflow coverage.
SubQ Ships First Commercial Subquadratic LLM With 12M Context
SubQ launched May 5 — first commercial subquadratic LLM, 1M API context, 12M research context, claimed 52x speedup at 1M, 1/5 cost of frontier models.
Googlebook Lands: Google AI-Native Laptop Replaces Chromebook
Googlebook unveiled May 12 — Aluminium OS, Magic Pointer, Create-your-Widget, Cast my Apps. Google bet on AI-native laptops ships fall 2026 via Acer, Dell, HP.
