AI Pulse - CW44

Hello Network! ??

To help navigate the rapidly evolving landscape of AI, I regularly curate industry updates for my teams, ensuring we remain informed about the most significant developments.

I’m pleased to now also share the most noteworthy AI news directly with you. ??

Super happy about every like ??, comment ?? or share ??!

Disclaimer: The external links provided in this article are for informational purposes only. I do not have any affiliation with neither the content nor the owners of these external websites and do not endorse responsibility for their accuracy, completeness, or reliability. Users are advised to review the terms and conditions and privacy policies of any external websites they visit.

?? Podcast Version of this Newsletter on Spotify ??

Disclaimer: From content curation to editing & voice synthesizing, AI has been an integral tool in streamlining parts of the production process. While every episode is still shaped by my own vision and input, this technology helps me deliver a seamless listening experience.

?? Listen In!

?? GitHub Copilot Updates & Spark Introduction at GitHub Universe 2024

GitHub Universe 2024 brought several exciting announcements regarding GitHub Copilot and the introduction of GitHub Spark.

GitHub Copilot Enhancements

GitHub Copilot, the AI-powered coding assistant, received significant updates:

Multi-Model Support:

• Copilot now offers a choice of AI models beyond OpenAI's. Users can select from Anthropic's Claude 3.5 Sonnet and Google's Gemini 1.5 Pro, in addition to OpenAI's models

Multi-File Editing:

• Developers can now edit multiple files simultaneously using Copilot.
• This feature allows for more complex code changes and refactoring across different parts of a project.

Xcode Integration:

• GitHub Copilot has expanded its support to include Apple's Xcode, facilitating AI-assisted coding for macOS and iOS developers.

Brainstorm Agent

• The Brainstorm agent is a new feature in GitHub Copilot Workspace that allows developers to have in-depth conversations about a repository.
• It's designed to help developers explore and understand unfamiliar codebases quickly.

GitHub Spark Introduction

The most groundbreaking announcement was the introduction of GitHub Spark, a new AI-powered tool for building web applications using natural language.

Natural Language App Creation:

• Spark allows users to create web applications by describing them in plain English, without writing code.

Accessibility:

• It aims to make app development accessible to non-coders, including product managers, designers, and anyone with an app idea.

Rapid Prototyping:

• Users can quickly build and iterate on app ideas through a chat-like interface.

Underlying Technology:

• Spark utilizes GitHub repositories, GitHub Actions, and Microsoft's Azure CosmosDB for database functionality.

Flexible AI Models:

• Users can choose between different AI models, including Claude Sonnet 3.5 and OpenAI's GPT models.

Sharing and Collaboration:

• Spark apps can be easily shared with customized access controls, allowing others to build upon the shared code.

Code Visibility and Editing:

• While Spark generates code automatically, users with programming knowledge can view and edit the underlying code if needed.

GitHub CEO Thomas Dohmke emphasized that Spark represents a significant step towards personal software creation, enabling users to explore ideas and build small helper applications without extensive coding knowledge.

These announcements reflect GitHub's commitment to expanding the accessibility of software development and leveraging AI to empower both experienced developers and newcomers to the field. It will be interesting to realize the limitations of no-code compared to pro-code as GitHub is pushing them closer together with the help of AI.

Full Keynote GitHub Universe 2024 (YouTube)

More on GitHub Spark (Blogpost)

?? OpenAI to build new chip venture with TSMC & Broadcom

OpenAI is making a significant move in the artificial intelligence hardware space by developing its own custom AI chip. Here are the key details about this initiative:

Collaboration with Broadcom and TSMC

OpenAI is partnering with two major semiconductor companies to create its first custom AI chip:

• Broadcom: Will assist with the chip design process
• TSMC: Will handle the manufacturing of the chips

This collaboration aims to have the first custom OpenAI chips ready by 2026, though the timeline remains flexible.

Motivations and Strategy

OpenAI's decision to develop its own chip is driven by several factors:

• Supply chain control: Reducing dependency on external suppliers like Nvidia
• Cost management: Addressing rising compute costs and operational losses
• Performance optimization: Creating chips tailored to OpenAI's specific AI model needs

Project Details

• A team of about 20 chip engineers has been assembled, including experts with experience designing Google's Tensor Processing Units (TPUs)
• OpenAI initially considered building its own chip manufacturing foundries but opted for design-only due to cost and time constraints
• The company is also diversifying its chip supply by incorporating AMD chips through Microsoft's Azure platform

This move by OpenAI follows a trend set by other tech giants like Amazon, Google, and Meta, who have also invested in custom chip design. OpenAI's venture into custom chip design represents a strategic effort to control its technological future and manage the escalating demands of AI computation. Owning the full vertical chain of the AI technology stack can be a key differentiator here.

News Article (Reuters)

?? Google Maps adding Gemini AI features

Google Maps is rolling out significant updates that integrate Gemini, Google's generative AI model, to enhance user experience and functionality.

AI-Powered Suggestions and Information

Google Maps is leveraging Gemini AI to provide more personalized and context-aware suggestions for places to visit and things to do.

Key Features:

• Users can ask open-ended questions like "things to do with friends at night" to get curated suggestions
• Gemini will offer AI-generated summaries of user reviews for recommended places
• Users can ask follow-up questions about specific aspects of a location, such as whether it has outdoor seating

These AI-powered features are rolling out this week in the United States for both Android and iOS users.

Enhanced Navigation and Route Planning

The update also brings improvements to navigation and route planning:

• Clearer display of lanes, crosswalks, and road signs on the map
• Precise lane guidance to help drivers navigate unfamiliar areas
• Option to explore landmarks, attractions, and dining options along your route before starting your trip
• Improved parking information, including nearby parking lots and walking directions to your final destination

Additional Features

• Weather-related road condition reporting, allowing users to see and report issues like flooding or low visibility
• Expansion of the Immersive View feature to 150 cities globally, offering 3D models of destinations

By integrating Gemini AI into Google Maps, the company aims to transform the app from a simple navigation tool into a more comprehensive guide for exploring and discovering new places. This is definitely a potential source for further revenue streams.

Blogpost (Google)

??? Further news

OpenAI's Sam Altman claims lack of compute is the reason for delaying products

OpenAI launches ChatGPT Search as a direct Google competitor

LinkedIn launches AI agent to replace job recruiters

Bye-Bye! ??

I appreciate you taking the time to read my take-aways & thoughts until here! If you enjoyed this week's article, please consider sharing.

Your engagement is super crucial to keep this newsletter running! ??

Cheers!

Robin