Scientists Train AI, AI Wins Nobel Prize

AI aces Nobel Prizes, new crazy chat-to-app capabilities, and more AI use-cases take the heavy lifting off of knowledge workers…

Nobel Prize Highlights AI’s Transformative Power

The 2024 Nobel Prizes just gave AI a standing ovation in Physics and Chemistry, recognizing groundbreaking work in neural networks and protein structure prediction.

These awards highlight AI's transformative role in advancing scientific understanding and accelerating breakthroughs across disciplines. It's almost like the Nobel committee is trying to tell us something about the future of science (hint: it's artificial and intelligent).

This is why they’re here:

• John Hopfield discovered a neural network that can memorize and reproduce patterns in various data types by processing and filling in missing information between data points. Now called the Hopfield Networks, they can be used to recognize images and optimize functions in computer science.

• Geoffrey Hinton's 1985 Boltzmann machine, inspired by Hopfield's work, learns from examples to recognize data characteristics and identify specific elements in images or patterns.

• Demis Hassabis, John Jumper, and David Baker won the Nobel Prize in Chemistry for using AI to predict and design protein structures. Their breakthroughs with AlphaFold and AI-driven protein design have transformed scientific research, making it possible to discover new pharmaceuticals and enzymes faster than ever before.

So What?

We can now see that these discoveries aren’t just shaping business—they’re flipping nature on its head. From mapping the universe of proteins to laying the foundation for neural networks, AI’s now solidified its place as “next up” in science, driving innovation like never before.

bolt.new Just Showed Us Coding is So 2022

Stack Blitz literally dropped a bomb on the AI community. We’ve seen Cursor, v0, and the Replit Agent, but bolt.new is a stunning combination of the three. 

Bolt.new is an AI-powered web development agent that allows you to prompt, run, edit, and deploy full-stack applications directly from your browser—with no local setup required.

The tools I mentioned before are great, but they operate in silos and you cannot run backends or directly edit code easily. That’s where bolt.new shines.

If you want some ideas to test its boundaries, prompt it with…

1. Create me a Discord clone where I can host a community.

2. Develop an application that allows me to record video using my camera while also sharing my screen, enabling me to talk through both at the same time.

3. Build an AI-based fitness tracking app.

Thought to app is absolutely becoming a thing, and fair warning, you’ll be blown away like I was. Give it a go for free 👇️ 

Gemini Handles Massive Case Loads for Google DeepMind ⚖️ 

AI is absolutely transforming the way companies manage routine tasks, including a pioneer itself, Google DeepMind.

Talk about augmenting workflows…

Google DeepMind’s legal team uses its Gemini AI model to handle up to 50% of information requests. According to DeepMind Director Terra Terwilliger, the AI helps the legal team quickly locate information they know exists but can’t immediately find, saving significant time on repetitive tasks and boosting overall efficiency.

Additionally, Bayer, like Google, uses AI to streamline tedious paperwork. The company’s AI automates 70-80% of the extensive regulatory filings required for its pharmaceutical research.

That’s a lot of documents off their plates!

So What?

While this seems straightforward enough, often the biggest wins in enterprise AI come from automating simple processes rather than full-on technological overhaul.

Training employees to use AI effectively requires time and effort, even for straightforward applications. Terwilliger emphasized that COOs should prioritize this learning process and reward employee engagement to ensure successful adoption. Investing in education will be key to maximizing AI’s benefits within organizations.

Perplexed by Perplexity? Here’s how to Use it Like a Pro

Perplexity AI is a tool that has been around for a while, but somehow it is still doesn’t get the flowers it deserves.

Perplexity stands out in the AI landscape by specializing in delivering accurate and real-time information, while models like ChatGPT and Google Gemini focus on creative, high-brainpower tasks. Here’s a breakdown:

• Creative tasks (e.g., brainstorming, editing, writing): ChatGPT and Google Gemini

• Accurate, real-time information: Perplexity and Google Search

For scenarios where factual accuracy is key—such as evaluating health supplements or troubleshooting software—Perplexity is a great option.

Here’s a quick guide how you can get started:

1. Navigate to Perplexity AI.

2. Start a new Thread.

3. Use key features like Focus and attaching your own PDFs to supplement your requests with additional context.

4. Integrate it into your workflow!

Our prompt picks to get you inspired…

1. Analyze the evolution of artificial intelligence in healthcare over the past decade. Discuss current applications, ethical considerations, and future prospects.

2. Develop a syllabus for a university-level course on the intersection of blockchain technology and sustainable development. Include key topics, reading materials, and project ideas.

3. Develop a comprehensive 30-day meal plan for a desk worker with specific dietary restrictions (gluten-free, dairy-free). Include nutritional information and sourcing tips.

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

Andrew Ng’s Deeplearning.AI just announced their new course on the new Llama 3.2 multimodal model from Meta.

Uber is set to launch an AI assistant, powered by OpenAI’s GPT-4o, to assist drivers with electric vehicle-related inquiries.

Microsoft has introduced Copilot Agents to OneDrive, offering custom AI assistants to automate tasks and enhance productivity.

Anthropic is introducing the new Message Batches API - a powerful, cost-effective way to process large volumes of queries asynchronously

OpenAI is beginning to lower its dependency on Microsoft data centers and secure its own compute capacity, with CFO Sarah Friar saying Microsoft hasn’t moved fast enough to supply the company with compute power.

The poll results from last edition were dead even, but that image IS NOT AI.

On a side note… We are launching our referral program today!

We want to give you a special gift for spreading the word about Digestible AI, so here’s the deal - refer just one friend to our newsletter and you will unlock access to our AI Tools Database; you can find it in the next section!

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

You can find our past newsletter editions here!

This newsletter was brought to you by Digestible AI. Was this forwarded to you? Subscribe here to get your AI fix delivered right to your inbox. 👋