Hey friend,

Today, we’ll explore

• How to automate your BlueSky content with N8N and AI.

• Claude's web search capabilities

• New audio models from OpenAI

• A free course for building AI agentic and multi-agent applications

Let's dive in.

How To Automate BlueSky With AI In N8N

This week, I published a tutorial showing how to use BlueSky within N8N. By default, there is no out-of-the-box solution for publishing content to BlueSky within N8N.

With the current political climate, BlueSky is a compelling alternative to X.

Check out this YouTube channel, where I release weekly video tutorials on building robust, production-ready AI automations using N8N and Python.

AI News

Here’s my pick of AI news this week:

1. Claude Gets Web Search Capability

Anthropic added web search capabilities to Claude, allowing it to surf the internet to find more current information. Available to paid US users as a preview, the feature uses Claude 3.7 Sonnet to deliver current information with direct citations for easy fact-checking.

Claude 3.7 is my favourite AI model by far so this is exciting to see.

• Claude's web search provides direct citations for easy fact-checking while delivering relevant sources in a conversational format, creating a more natural and reliable information retrieval experience.

• The new search capability expands Claude's knowledge base with real-time information and improves accuracy on tasks requiring recent data, addressing a key limitation of previous AI assistant versions.

• Available as a preview for paid US users, the feature enhances professional workflows for sales teams analysing trends, financial analysts assessing markets, and researchers building literature reviews.

2. Google Gemini Introduces Canvas and Audio Overview Collaboration Tools

Google has released two major collaboration features for Gemini: Canvas, an interactive space for real-time document and code creation, and Audio Overview, which transforms documents into podcast-style AI discussions. These additions enhance Gemini's capabilities as a creative and collaborative assistant across text, code, and audio formats.

• Canvas provides an interactive environment where users can generate drafts, receive AI feedback, and preview HTML/React code with real-time changes, exporting finished work to Google Docs for further collaboration.

• Audio Overview generates conversations between two AI hosts that summarize uploaded documents, slides, and research reports, with options to share or download audio for offline listening.

• Both features are rolling out globally to Gemini subscribers, positioning Google's AI as a more versatile assistant for writers, developers, and researchers while streamlining creative and professional workflows.

3. OpenAI Releases Next-Gen Audio Models for Enhanced Voice Applications

OpenAI has released new audio models for speech-to-text and text-to-speech capabilities through its API. The models—gpt-4o-transcribe, gpt-4o-mini-transcribe, and gpt-4o-mini-tts—set state-of-the-art transcription benchmarks while enabling developers to create more intelligent and customizable voice agents.

• The new models are built on GPT-4o architectures using specialised audio-centric datasets, advanced distillation techniques, and reinforcement learning to achieve superior transcription accuracy across diverse accents, noisy environments, and varying speech speeds.

• GPT-4o-mini-tts offers enhanced "steerability" through natural language instructions, allowing developers to control speaking styles with commands like "speak like a sympathetic customer service agent" for more expressive and contextually appropriate AI voices.

• Available immediately to all developers via OpenAI's API, these models support applications ranging from customer service and meeting transcription to expressive narration and voice agents, advancing OpenAI's broader initiative to create more capable AI assistants.

What I Found Interesting This Week

I found this great free course about how to use LangGraph. LangGraph, created by LangChain, is an open source AI agent framework designed to build, deploy and manage complex generative AI agent workflows. It provides a set of tools and libraries that enable users to create, run and optimise large language models (LLMs) in a scalable and efficient manner.

Check it out here. I’ll be diving in this week.

Your Opinion Matters

What did you think of today’s email? Your feedback helps me create better emails for you!

Got more feedback or want me to cover a specific topic? Reply to this email and let me know.

Owain