We are starting to see what "AI will accelerate science" actually looks like.
This Google paper describes novel discoveries being made by AI working with human co-scientists (something I think we have all been waiting to see), along with an early version of an AI scientist.
There is a lot of important stuff in this new paper by Anthropic that shows how people are actually using Claude.
1) The tasks that people are asking AI to do are some of the highest-value (& often intellectually challenging)
2) Adoption is uneven, but many fields already high
🚨 New @a16z thesis: building websites / apps with AI
There's been an explosion of products that help users "vibe code" a web app from text prompts.
We dove deep on these tools - who's using them, how they work, and where they might be headed.
Our market map + insights 👇
3...2...1...🚀 blast off into the era of AI agents!
We've gathered 321 examples of gen AI having a real world impact across industries. Dive into the blog and see what our customers have been building ↓ https://t.co/bm6j3nM25Q https://t.co/rn7UAw65Ww
🪄 LangChain State of AI 2024
What LLMs are the most widely used today? What metrics are commonly used for evals? Are developers finding success in building agents?
Our State of AI 2024 report shows where the AI ecosystem is headed, based on data from LangSmith. Key 5 insights in the thread 🧵👇
Full report: https://t.co/J8Aokheh1K
The new Google Gen AI SDK provides a unified interface to Gemini 2.0 through the Gemini API. Check out the cookbook to get started ↓ https://t.co/CAzvQRaNy2
📢 Releasing Agent Poirot. 🚀
It is a Data Analytics Agent that can extract actionable, tailored insights that go beyond basic statistics.
It achieves SOTA on the insight-bench analytics benchmark [1].
Github: https://t.co/2HaWKdqCyV
Paper [1]: https://t.co/kqIUx98W9T
Excited to share that the AlphaFold 3 model code and weights are now available for academic use.
Looking forward to seeing what new research this unlocks and how the research community builds on AlphaFold 3 for scientific discoveries https://t.co/GKIOGHm317 1/2
Our industry-leading AI search model is now multimodal!
Embed 3 enables enterprises to build systems that can accurately and quickly search across both text and image data sources like complex reports, product catalogs, and design files.
https://t.co/wsW0u6chcy
"These effects are large. To put the rise in materials discovery in perspective, the lab’s research output per scientist declined by 4% over the preceding five years.
This was despite the introduction of several computational tools designed to aid scientists.
AI therefore appears to be a different class of technology, with impacts that are orders of magnitude greater than previous methods." 👀
Today, we're announcing that @fastdotai is joining @AnswerdotAI, marking a new phase in making AI accessible.
And we're launching a new a new kind of "AI-first" educational experience, "How To Solve It With Code".
https://t.co/hLQ4gAZnsz
Our new AI paper reveals surprising geometric structure in the LLM-learned concepts: 1) They form brain-like "lobes", 2) they form "semantic crystals" much more precise than it first seems, and 3) the concept cloud is more fractal than round:
🌉The bridge in your hometown
🏞️The natural park in your region
🏰The cultural heritage castle in your city
You can now explore over 1.9 million EU-supported projects on our new portal!
Easily navigate and see the impact of EU funding across Europe! ↓
https://t.co/4OOzUByCJ6
Shoutout to the team that built https://t.co/sJnTRDfHGF . Really neat site that benchmarks the speed of different LLM API providers to help developers pick which models to use. This nicely complements the LMSYS Chatbot Arena, Hugging Face open LLM leaderboards and Stanford's HELM that focus more on the quality of the outputs.
I hope benchmarks like this encourage more providers to work on fast token generation, which is critical for agentic workflows!