This past week we hosted @AnthropicAI's inaugural Builder Summit in Paris on the heels of the AI Action Summit. While the broader discussions this week was on global governance, the focus of our activations were on the builders actually taking action shaping how AI manifests in

Anthropic and Lovable co-hosted an exclusive hackathon at Slush 2025 called Push to Prod. 100+ builders competed to build production-ready apps in 60 minutes. The winners @0xSorcier @TGuidoux @ClementGarcia_ built @cliccc_ai and won dinner with Anthropic, Lovable, and VCs, and

In 2000, the S&P 500 had an elevated trailing P/E of 26, well above its long-term avg. of 14. For the next 11 years, the S&P 500 delivered zero returns! The current S&P 500 trailing P/E is 30 – even higher than 2000. History rhymes. For standardized performance visit our website.

Brilliant move by @AnthropicAI to sponsor Claude ads on stacktraces that get no results

@Lovable is a great example - they launched at Slush 2024. One year later: $200M ARR. We wouldn't be surprised if at @SlushHQ 2026, the next breakout startup keynoting mainstage is one of the founders who pushed their idea to production this past week. If you have an idea

The intelligence part of Claude only became more powerful this past week when we launched Opus 4.5 and we saw this on the ground.  Technical founders make infrastructure decisions in hours, not months- they need credits and an API key and they’re off to the races.

There's a spirit of revival stirring in America. Dinesh D'Souza's new film, The Dragon's Prophecy, is part of that movement. Based on Jonathan Cahn's No.1 bestseller, the film reveals how war, politics and the Bible converge in a way that we need to make a choice. Watch on GJW+.

Five finalists pitched to a panel of Lovable and Anthropic leadership, @zhenyaloginov from @Accel and @Kitty_Mayo_ from @ProjectEurope. Demos ranged from Craci's EU Cyber Resilience Act compliance automation to Postable's AI-powered brand profile generator to Lavoe's

We also hosted back-to-back technical workshops for 2000+ founders: @HoagyCunningham went deep on Interpretability for Trust, taking founders inside the "black box" of neural networks to show how Anthropic is building trust into increasingly capable AI systems @The_Whole_Daisy

In a 2-day whirlwind we celebrated @Lovable's $200M ARR milestone at their first birthday party, followed by our flagship joint hackathon and a @Lovable x @AnthropicAI x @BessemerVP dinner for our winners. Together with @antonosika and @FabianHedin from Lovable and @G_Princen

Dario's essays and long debate slack threads are one of my favorite parts of Anthropic's culture. They're open, detailed - and incredibly raw. Everyone at the company ends up having a good sense of how the company is making decisions and what matters. Its the kind of thing that

Claude Opus 4.5 is our best model yet. The best in the world at coding, agentic tasks, and everyday work like spreadsheets. Engineers keep telling us Opus 4.5 “just gets it.” Let us know what you think!

"Expert graders compared deliverables from leading models to human experts. Claude Opus 4.1 produced outputs rated as good as or better than humans in just under half the tasks." Source: https://t.co/CelcOAOshK

Link: https://t.co/CelcOAOshK “Claude Opus 4.1 achieved a 47% win-plus-tie rate. What does that mean? In nearly half of all tasks, expert graders rated Claude's work as equal to or better than the work produced by human professionals.” “This was tested across 220 tasks spanning

wait did OpenAI just say “Claude is best” in their own blogpost 🤔

We disrupted a highly sophisticated AI-led espionage campaign. The attack targeted large tech companies, financial institutions, chemical manufacturing companies, and government agencies. We assess with high confidence that the threat actor was a Chinese state-sponsored group.

Lovable and Anthropic are hosting an AI hackathon. It’s called Push to Prod, happening at Slush in Helsinki and I’ll be there! Apply to participate:

Next week, the @AnthropicAI startup team will be in Korea. A similar pattern is emerging there. The best builders in every market are realizing the same thing: we're not in the "pick a winner" phase of AI. We're in the "compound your velocity" phase. The founders who win in

Major thanks to our partners -  Nicolas Ishihara and @jrsyo from @Vercel, Stephanie Leong, Pedro Duarte, and @thomaspaulmann from @Raycast, @dshukertjr from @supabase, @vabgoel from @nttventurecap, @l2k from @wandb - we are so grateful to you all and excited for future collabs!

We saw this first hand at our Claude Code Hackathon where devs discovered they could build in 3 hours what would've taken 3 months. One team built a @raycast extension that turned natural language into complex database queries. Another automated their entire deployment

The Japanese startup ecosystem understands this urgency at its core: @HidetoshiTojo-san, our Head of Japan highlighted in a panel with CXOs from @Rakuten, @moneyforward and @classmethod showed how Claude Code is accelerating software development: 👨‍💻  @Rakuten - 7 hours of

The overwhelming takeaway from the founders we met is that what matters the most in 2025 is not betting on a single model. When you partner with @AnthropicAI, you know you’re building on top of a frontier AI lab that deeply understands founder mode at its core and ships as

This past week we officially launched @claudeai in Japan and entered the market with an eye on startups - kicking off with our Tokyo Builder Summit, and joining forces with @vercel, @raycast, and @supabase for a Claude Code Hackathon, before capping the week off with a founder