The $17M Bet on AI Web Navigation

Browser Use has officially captured the attention of investors…

In this edition we’ll be covering…

• The latest developments on Browser Use

• A tutorial on how to use the new Google Gemini Canvas

• A breakdown on OpenAI’s new audio models

• 5 AI trends

• 3 AI tools to keep you productive

• And much more…

The Latest in AI

The $17M Bet on AI Web Navigation

Navigating the web has always been a breeze for us humans—click, scroll, repeat.

But for AI agents? Not so much. Enter Browser Use, a startup that’s making the internet more digestible for our silicon-based friends, and they’ve just bagged a cool $17 million in seed funding to prove it.

What’s the big deal?

• Browser Use transforms complex website elements into a structured, text-like format, enabling AI agents to understand and interact with web pages more effectively.

• Founded by Magnus Müller and Gregor Žunič at ETH Zurich, the project quickly gained traction after its open-source debut, showcasing the power of community-driven innovation. We covered how to use the self-hosted version here.

• With Felicis leading the funding round and participation from notable investors like Paul Graham, A Capital, and Nexus Venture Partners, it’s clear that the smart money sees potential here.

So What?

As AI agents become more prevalent, their ability to navigate and interact with the web seamlessly is crucial. Browser Use’s technology could serve as a foundational layer, ensuring that AI can “read” the web as effortlessly as we do.

It’s like giving AI a pair of reading glasses for the internet—because nobody likes a squinting robot…

If you don’t want to go through the hassle to self-host Browser Use, you can use the cloud version (it’s paid)! 👇️ 

Get Your Hands Dirty!

Canvas comes to Gemini in Style

Gemini Canvas just killed context switching.

Similar to ChatGPT, Gemini now has an interactive workspace that lets you refine documents and code, without switching tabs.

Some use-cases include making an infographic interactive, adding a frontend to your existing backend code, and enhancing creative writing.

This brings Gemini much closer to an end-to-end productivity environment, especially for:

• Content Creators

• Developers

• Knowledge workers

Here’s how you can use it yourself to generate a landing page:

1. Navigate to Gemini.

2. Under the chat window, toggle on Canvas.

3. Use this prompt:

I need help creating a quick landing page for my website. I'm looking for a clean and modern design. The purpose of the landing page is to [State the goal: e.g., collect email sign-ups, promote a new product, drive event registrations]. My target audience is [Describe your target audience]. I'd like to include the following sections: [List desired sections: e.g., hero section with headline and call to action, a brief description, a benefits section, a sign-up form/button]. Can you help me with ideas for headlines, copy, and a basic layout?

Personally, I find these Canvas features so underrated. Loving this from Google so far!

Industry Intel

Talk to Me Nice (Says OpenAI)

OpenAI is turning up the volume on AI interactions with the release of advanced speech-to-text and text-to-speech models.

Imagine instructing an AI to “speak like a medieval knight” and hearing it respond in perfect chivalric prose. OpenAI’s latest audio models are turning this into reality.

• The newly introduced gpt-4o-transcribe sets a new benchmark in speech-to-text accuracy, excelling in understanding diverse accents, filtering through noisy environments, and adapting to varying speech speeds.

• With OpenAI’s latest text-to-speech model, gpt-4o-mini-tts, developers can instruct the AI not just on what to say, but how to say it.

• To experience these breakthroughs firsthand, check out OpenAI.fm. It’s a demo site where anyone (technical or non-technical) can test these new voice capabilities!

So What?

OpenAI’s advancements in audio modeling are not just technical feats; they herald a new era in human-computer interaction. The ability to customize AI voices to such a granular degree opens up unprecedented possibilities for personalized user experiences.

As these models become more integrated into everyday applications, expect interactions with AI to feel more natural, engaging, and, dare we say, human…

Quick Bites

Stay updated with our favorite highlights, dive in for a full flavor of the coverage!

TikTok’s owner ByteDance has less than two weeks to sell the social media platform. Now, Perplexity is making the case for why it should become the popular video-sharing service’s new owner.

Zapier just launched MCP support to to let your AI assistant interact with thousands of apps.

Google has started rolling out new AI features to Gemini Live that let it “see” your screen or through your smartphone camera and answer questions about either in real-time.

A Texas private school is seeing student test scores soar to top 2% in the country following the implementation of an AI "tutor."

Alibaba affiliate Ant Group is using both Chinese and U.S.-made semiconductors for building more efficient AI models.

Trending Tools

🎙️ Suno - an AI-powered music generator that allows users to create original songs by simply describing the type of music they want.

✈️ Starter Pilot - A full AI-powered kit to take your idea from 0 to 1.

📗 Automateed - Full tool to generate AI ebooks.

The Neural Network

Many of you expressed interest in Model Context Protocol in last week's poll, and we hear you!

We'll be featuring comprehensive MCP tutorials in upcoming editions, but for those eager to get started, here are some excellent resources:

• Get Started with Model Context Protocol in 2 Minutes from Matt Pocock

• Building Agents with Model Context Protocol - Full Workshop with Mahesh Murag of Anthropic

• Model Context Protocol (MCP) - Explained by Marco Codes

Building something cool with MCP? We'd love to see it, so share your projects with us!

Until we Type Again…

Thank you for reading yet another edition of Digestible AI!

If you have any suggestions or specific feedback, simply reply to this email. Additionally, if you found this insightful, don't hesitate to engage with us on our socials and forward this over to your friends!