Good morning. It’s Monday, October 30th.

Did you know: OpenAI is set to discuss new tools at its upcoming DevDay conference on November 6th, which will stream live at 10:00am.

In today’s email:

• GPT-4 Updates

• Corporate Investments & Collaborations

• Product & Tool Updates

• 5 New AI Tools

• Latest AI Research Papers

You read. We listen. We’re going more in-depth on some stories, as well as providing more context to the recommended AI tools. Let us know what you think of this edition by replying to this email, or DM us on Twitter.

GPT-4 Upgrades: PDF support, “All Tools” available in one chat

OpenAI has launched a new version of ChatGPT that integrates all GPT-4 tools, allowing users to access features like advanced data analysis, DALL·E 3, and built-in browsing - without needing to switch between them.

The updated ChatGPT now supports the analysis of various document types, including PDFs. The "All Tools" feature enhances user experience by providing a unified platform, potentially making several third-party ChatGPT plugins redundant. OpenAI's move aims to offer a more powerful tool based on user feedback without depending on external functions. Furthermore, some users have observed an updated knowledge cutoff date in the new version.

This feature is currently on a ChatGPT Plus rollout, and users may notice these features available in their premium subscription within the next few days.

To follow the latest release notes from OpenAI, you can bookmark their Release Notes page.

A Big Week for Investments

Google agrees to invest up to $2 billion in OpenAI rival Anthropic with a $500 million immediate investment, with the remainder to follow. Amazon also said last month it would invest up to $4 billion in Anthropic to compete with the Microsoft + OpenAI powerhouse.

Speaking of investments, what are the top performing publicly-traded AI companies this year?

New AI Industry Body: The Frontier Model Forum

Chris Meserole has been appointed as the first Executive Director of the Frontier Model Forum, an industry body aiming for the safe and responsible development of frontier AI models.

The announcement came jointly from Google, Microsoft, OpenAI, and Anthropic. Additionally, these companies, along with philanthropic partners such as the Patrick J. McGovern Foundation, the David and Lucile Packard Foundation, Eric Schmidt, and Jaan Tallinn, have committed over $10 million to a new AI Safety Fund to support research into AI safety and will also back independent researchers globally.

The Frontier Model Forum is focusing on sharing best practices on "red teaming" and is also working on a responsible disclosure process for AI vulnerabilities.

OpenAI Leases Massive SF Office from Uber

OpenAI inks a massive San Francisco office lease of two buildings from Uber's Mission Bay headquarters in San Francisco at 1455 and 1515 Third St.

This represents the largest office lease in San Francisco since 2018. The lease covers almost half a million square feet, allowing OpenAI to accommodate over 2,000 employees. Other startups, such as Anthropic, Hive AI, and Hex, have also recently acquired spaces in San Francisco, particularly in "Area AI" which spans Potrero Hill and Showplace Square.

Interested in exploring “Area AI” up close? Check out Google’s new Immersive View using neural radiance fields for detailed route visualization.

“AI Exposure” is the new buzz term for job loss

The rapid growth of AI has heightened concerns over job displacement, especially in knowledge worker roles, which are cognitively-oriented.

While past technological advancements primarily threatened manual jobs, now white-collar jobs are also at risk. Even when LinkedIn laid off 700 workers, discussions shifted to AI's role in job losses, though there was no direct indication from LinkedIn about AI's involvement. Alan Guarino of Korn Ferry highlighted that AI challenges cognitively-valued jobs and could either be a displacement threat or a tool for enhanced work efficiency.

In May, 4,000 U.S. jobs were lost due to AI. Mentions of generative AI in corporate discussions surged from five in December 2022 to 390 in June 2023. A Rand Corporation study found that all U.S. occupations are exposed to technological advancements, with those requiring higher education being increasingly at risk.

1. FlowRL - Creates Unique App UI Based on Behavior

flowRL provides AI-powered real-time UI customization. Advanced ML tailors app experiences to user behavior, boosting retention and revenue. Eliminates lengthy A/B testing, and streamlines feature development. Achieve growth with personalization.

2. GetMax - AI Content Generator For Small Businesses

GetMax is an AI-driven content marketer, optimizing strategies at a fraction of traditional costs. It identifies customer profiles, crafts tailored strategies, automates content creation, and offers one-click posting. Ideal for startups seeking efficient, cost-effective content solutions.

3. Penseum - Create Study Material with AI

Penseum is a 24/7 personalized learning tool, transforming notes, PDFs, videos, and slides into study aids. It features custom guides, flashcards, Q&As, AI chat, a study roadmap, and an AI tutor. Users can join a 1000+ Discord community for shared experiences.

4. Chatmind - AI Generated Mind Maps

Chatmind offers chat-driven mind mapping and brainstorming, enabling users to create and adjust maps conversationally. Features include auto-saving, easy edits, theme changes, and upcoming AI capabilities. Users receive free credit upon sign-up.

5. Olle AI - Built-in AI Assistant for Mac Apps

Olle AI is a macOS toolbar enhancing productivity via AI. It features offline prompts, dynamic creation, AI chat, seamless app integration, and customizable output. Prioritizing privacy, all data stays local. One-time $25 download price, no subscriptions!

📄 CODEFUSION: A Pre-trained Diffusion Model for Code Generation 

📄  FP8-LM: Training FP8 Large Language Models ​​

📄  ZEPHYR: Direct Distillation of LM Alignment

📄 PockEngine: Sparse and Efficient Fine-tuning in a Pocket

📄  Data-Centric Financial Large Language Models

Thank you for reading today’s edition.

Your feedback is valuable.

Respond to this email and tell us how you think we could add more value to this newsletter.