💰Anthropic's $183B Valuation After Massive Funding
Anthropic just announced a new $13B funding round that values the company at a whopping $183B, nearly tripling its worth from six months ago and coming amid significant growth from Claude Code and enterprise accounts.
- —Anthropic now serves over 300,000 enterprise customers, with accounts worth $100k+ in annual revenue growing sevenfold in 2025
- —The company's Claude Code assistant now generates $500M in annual revenue, with usage up 10x in three months
- —The investor list notably includes QIA, the first Middle East participant, as hinted at by CEO Dario Amodei in a leaked memo in July
- —Anthropic's last raise of $3.5B in March valued the company at $61.5B, marking a meteoric 3x rise in the last six months
Why it matters: While there has been plenty of chatter about an AI investment bubble, the dollars don't seem to be drying up yet, especially for leaders on the frontier of the industry. While Anthropic previously turned down investments from the Gulf region, the expensive race for global compute from rivals may have changed some perspectives.
📊OpenAI Acquires Statsig for $1.1B
OpenAI just announced a $1.1B all-stock acquisition of experimentation platform Statsig, bringing its founder Vijaye Raji aboard as CTO of Applications to lead ChatGPT and Codex engineering.
- —Raji will oversee product engineering across ChatGPT and Codex, reporting to Applications CEO Fidji Simo, who recently joined from Instacart
- —Statsig's platform, which powers A/B testing and feature flagging for companies including OpenAI itself, will continue operating independently
- —The deal marks OpenAI's second major acquisition attempt this year, following a failed $3B bid for coding startup Windsurf in July
- —OAI also made several other executive moves, with Srinivas Narayanan shifting to CTO of B2B Apps and CPO Kevin Weil leading a new "AI for Science" team
Why it matters: OpenAI's last acquisition was a bumpy ride, but the leadership movement appears to have this one on solid ground. With the competition for talent at an all-time high in the AI race, the acqui-hire route may become an even more utilized tool for the giants of the space to bring in top leaders and teams in areas of need.
🍌Create Product Ads with Nano Banana
In this tutorial, you will learn how to use Google's new Nano Banana model (Gemini 2.5 Flash Image) to turn product photos into professional marketing ads and transform sketches into animations — all for about 4 cents per image.
- —Go to gemini.google.com, click "Tools" and toggle on "Create Images" (that's Nano Banana/Gemini 2.5 Flash Image)
- —Upload your product image and prompt: "Create ads to fit various environments with the attached image — subway, train station, bus stop, luxury magazine"
- —Watch as it seamlessly integrates your product into realistic ad environments with appropriate lighting and perspective
- —Transform sketches by uploading black and white drawings — Nano Banana adds color while preserving exact character structure
- —Take it further: Toggle on "Create videos with Veo" to turn your generated images into animated videos with sound (takes ~2 minutes)
Why it matters: Mock up entire campaigns before spending real money — at 4 cents per image through the API, test different concepts on social media to see what resonates before committing to expensive production.
🌎Tencent's Voyager for 3D World Creation
Tencent just released HunyuanWorld-Voyager, an open-source "ultra long-range" AI world model that transforms a single photo into an explorable, exportable 3D environment.
- —Voyager uses a "world cache" that stores previously generated scene regions, maintaining consistency as cameras move through longer virtual environments
- —It topped Stanford's WorldScore benchmark across multiple metrics, beating out other open-source rivals in spatial coherence tests
- —Users can control camera movement through keyboard or joystick inputs, with just a single reference photo needed to create the exportable 3D environments
- —The system also remembers what it creates as you explore, so returning to previous areas shows the same consistent scenery
Why it matters: World models have become one of the hottest frontiers in AI, with labs racing to build systems that understand physical spaces rather than just generating flat images. Between Genie 3, Mirage, World-Voyager, and more, the range of options (and the applications for these interactive 3D environments) is growing fast.