BB Digest: Claude learns to click & type

Digest #504 → Subscribe | Upgrade to Pro

Hey folks, I have some updates to share:

1. Our lifetime pricing for Ben’s Bites is ending very soon 😲 . If you haven’t taken the leap yet, you still have a little time - sign up here for $250 (you can expense it). We’ve got 370+ tutorials and courses, plus season 2 of our workshops is coming soon, more content gets added every week.

2. Our free AI Marketing program is blowing up! 300+ marketers have already signed up. It’s a 3-week program starting November 4th (async, with some live sessions). Register here to join.

3. We're also creating a guide on AI in Sales and looking for interesting case studies. Are you using AI for sales in a unique or successful way? We'd love to feature your story - fill out this quick form.

4. and, I just added a new workshop for 31st Oct with Tom on how to build AI voice assistants without code.

TLDR; inside today’s newsletter

• Claude learns computers, Ideogram vs Midjourney's image editors, and new creative AI tools

• snack n skill - A bite-sized lesson to analyze competitor reviews with AI

• 6 new tools for documentation, video editing and more

• 8 interesting posts covering open-source models to enterprise AI

• to-do’s

Alright, let’s get to it!

 🔎 Trends & news

• Claude learns to use computers - Anthropic released a “New” version of Claude 3.5 Sonnet with a) a massive boost to its coding powers (49% on SWE-bench, up from 33%) and b) the ability to navigate interfaces and use software like humans do (in beta). We have word on Claude 3.5 Haiku too but Opus is missing. Computer Use is primarily usable by developers for now, but you can read notes from Ethan and Simon on how it changes the game.

• Ideogram launches Canvas - An infinite creative space with Magic Fill for inpainting and Extend for outpainting, perfect for combining uploads with AI generations. Not to be outdone, Midjourney has revealed an image editor plus re-texturing tools for materials and lighting.

• Join over 200,000 developers who trust AssemblyAI's robust API and industry-leading accuracy to build next-gen AI products. Experience the most advanced speech-to-text models—with up to 95% accuracy. Get $50 in API credits when you sign up today!*

• Run AI inference on the fastest GPUs or fine-tune the latest models on Modal's serverless compute platform. Only pay for what you use. Run a GPU job today in seconds directly from our brand new Modal playground. Startups and researchers eligible for up to $25k in credits*.

• AI for multimedia isn’t stopping because RunwayML also announced Act-One in Gen-3 Alpha. It turns simple video performances into expressive character animations without mocap equipment.

  ps: more creative tools and news below.
  *sponsored

🍽️ Snack ‘n’ skill

Learn how to analyze competitor Google reviews using AI

Stop spending hours manually researching competitor reviews. Here's how to automate gathering and analyzing Google reviews to understand what customers think of your competition.

Step 1. Build your listening post

Create an Airtable base with essential fields: business keywords to search, location names, and a status field to trigger the automation. This will be your command center for competitor review analysis.

Step 2. Set up your insight engine

• Connect Airtable to Zapier and trigger actions when status changes to "Go"

• Use SerpApi to fetch Google Maps locations and their reviews

• Set up a loop in Zapier to process multiple locations

• Feed reviews through ChatGPT for intelligent summarization

• Automatically update Airtable with location details and AI-generated summaries

• Use the GPT-4-mini model to keep costs low while maintaining quality

Tip: Start with your closest competitors first. Once you've refined the AI's summary style, you can expand to analyze the whole market.

This walkthrough is a condensed version of this tutorial. Check it out for step-by-step API setup details and one-click copyable prompts.

⚙️ Top new tools

• Guidde* - Create stunning visual guides in seconds—the free GPT tool for fast, AI-powered documentation. Save time and effort!

• DreamCut - AI video editor and screen recorder that works right from your browser.

• Perplexity Pro now has a reasoning mode to answer multi-layered questions.

• Voice Design by ElevenLabs - Generate a unique voice from a text prompt alone.

• Watermark PDF - Add custom text or image watermarks to your PDF documents securely and easily.

• Together Demos - Open source example apps by the Together AI team.

More tools here →
*sponsored

 📜 Interesting posts

• Learn to use Generative AI as a Product Manager with this free course only for Ben’s Bites readers.

• Enterprise AI insights from industry leaders (from SF Tech Week).

• Appen's 2024 State of AI report - Generative AI and its impact on business processes.

• Genmo AI open-sourced their (amazing) video generation model Moshi-1 preview.

• Inflection AI partners with UiPath to bring agentic workflows to their enterprise plans.

• Even xAI is hiring to build autonomous agents.

• Google Deepmind open-sourced Synth-Id and has made significant improvements to MusicFX.

• OpenAI has shared new research that allows them to create images faster.

📌 To-dos

1. Let us know if you have any tutorials or courses you want us to create—we’re always open to ideas.

That wraps things up for today! See you again next week. 👋

Ben