- The Rundown AI
- Posts
- OpenAI reveals insider emails
Welcome, AI enthusiasts.
The OpenAI vs. Elon Musk saga just took another twist — with Sam Altman, Ilya Sutskever (!) and co. pulling out email receipts to refute the billionaire's claims.
But if Musk's follow-up tweets are any indication, this fight is far from over. Let’s investigate…
In today’s AI rundown:
OpenAI fires back at Musk with receipts
Whistleblower sounds alarm on Copilot Designer
Run Google’s latest LLM on your phone
AI math tutor boosts performance in Ghana
5 new AI tools & 4 new AI jobs
More AI & tech news
Read time: 4 minutes
LATEST DEVELOPMENTS
OPENAI
Image source: OpenAI
The Rundown: OpenAI just punched back at former co-founder Elon Musk’s lawsuit, releasing a blog post containing a series of his own emails to refute claims that the company abandoned its original mission in pursuit of profits.
The details:
Emails show Musk urging OpenAI to raise at least $1B, pushing the nonprofit to find more financing to achieve its lofty AGI goals.
Musk also suggested in 2018 that Tesla should acquire OpenAI as "the only path" for the startup to stay relevant and beat Google.
Also of note is Ilya Sutskever, listed under the blog’s authors after being absent from public view since November’s drama.
Sutskever said in one email that the ‘Open’ in OpenAI refers to everyone benefitting from AI, not sharing the science — raising eyebrows in the open-source AI community.
Musk responded to the blog by tweeting that he would drop the lawsuit if the company changes its name to ‘ClosedAI’.
Why it matters: The drama surrounding OpenAI continues to be the best reality show in tech, and Elon’s latest cameo adds a whole new level of spiciness. While the emails tell another side of the story, Musk is not one to back down from a fight — meaning there are likely more fireworks to come.
TOGETHER WITH GALILEO
The Rundown: Learn how to build GenAI apps from top GenAI experts from LinkedIn, Google, Coinbase, Roblox and more.
At GenAI Productionize, you’ll learn:
The emerging enterprise GenAI stack, from orchestration to evaluation, inference, retrieval, and more
Practical hands-on architecture and tool insights
How organizations govern, evaluate, and monitor GenAI
MICROSOFT
Image source: Getty Images
The Rundown: A senior software engineer at Microsoft just escalated concerns to the FTC regarding Copilot Designer, warning the company is not doing enough to prevent the AI from generating harmful content.
The details:
Shane Jones discovered a vulnerability in OpenAI’s DALL-E (which powers Copilot Designer), allowing him to bypass content filters.
Jones said he urged the company to pull the product until more safeguards were in place, but was ignored.
In his letter, Jones alleges the tool depicts graphic violence, underage drug/alcohol use, political bias, copyright violations, and more.
Why it matters: This explosive complaint from within Microsoft's own ranks seriously undermines the tech giant's public commitment to "responsible AI" principles — while also suggesting profits may be trumping safety.
AI TRAINING
The Rundown: In this tutorial, you will learn how to run Google’s latest Gemma 2B model on your phone, which offers privacy and access to it without an internet connection.
Step-by-step:
For iPhone users, there is currently a beta app.
Click here, where you’ll be shown two steps. First, select step 1, “View on App Store,” and download Testflight.
Go back to the website and click on step 2, “Start Testing” to install MLC-Chat.
Once you install the app, press the “Cloud” icon next to the Gemma 2B model to download. You can also download and use other models such as Mistral 7B Instruct or Phi-2.
Wait a few seconds, and you’ll be able to chat with it.
For Android users, you can find a demo app as well, where you will have to download the APK file.
PRESENTED BY ATHENA
The Rundown: 100s of top performers use Athena's elite executive assistants to maximize time. Athena hand picks the top 1%, continuously trains on the latest Al tools, and helps people 10x productivity.
Delegate tasks like:
Getting to inbox 0
Generating high-quality leads
Growing your personal brand
Get $1500 towards your Athena EA. Exclusive to The Rundown readers until March 31st.
AI & EDUCATION
Image source: University of Oxford
The Rundown: A new study in Ghana found that students who used Rori, an AI-powered chatbot math tutor, achieved significantly higher math scores compared to just regular math lessons.
The details:
Oxford researchers divided 1,000 students into two groups, with one using Rori for two sessions per week in addition to normal math lessons.
After 8 months, the Rori students had significantly higher math growth scores than the control group.
Rori works on mobile devices and low-bandwidth networks.
The projected cost per student is also just $10.
Why it matters: Massive improvements in student performance from a simple, low-cost AI intervention are an exciting example of the tech’s potential. AI may just be the turning point in unlocking world-class education for millions of disadvantaged students globally.
NEW TOOLS & JOBS
🔵 Circleback- Unbelievably good AI-written meeting notes and powerful automations with Notion, Slack, HubSpot, and more. Try for free.*
💬 Chat with MLX- MacOS application for efficient document interaction
👗 DeepFashion- Create personalized looks with AI-generated fashion images.
🤖 Ema- A trusted AI employee automating complex workflows
👀 Dopt- In-app assistant that unblocks users
*Sponsored listing
QUICK HITS
Section is hosting a free workshop on AI for Personal Productivity on March 14. In two hours, learn to set up your "AI workspace," identify AI use cases, and build a habit that sticks. RSVP here.*
Palantir won a $178M contract from the U.S. Army to deliver 10 prototypes of the TITAN vehicle, which uses AI and data fusion to provide soldiers with advanced battlefield intelligence capabilities.
Amazon announced a $53M investment into Glacier, an AI startup working to improve recycling efficiency and reduce waste.
Google unveiled new changes to its search ranking algorithm to combat the rise of AI spam and reduce low-quality content in results by up to 40%.
Ebay is building its own LLM trained on its e-commerce data, with plans to double its GPU capacity over the next year.
Anthropic researcher Amanda Askell posted a breakdown of Claude’s system prompt, revealing a simple, concise structure with an August 2023 knowledge cutoff date.
Cohere announced a new partnership with Accenture, collaborating to accelerate enterprise adoption with customized AI solutions.
*Sponsored Listing
THAT’S A WRAP
SPONSOR US
Get your product in front of over 500k+ AI enthusiasts
Our newsletter is read by thousands of tech professionals, investors, engineers, managers, and business owners around the world. Get in touch today.
FEEDBACK
How would you rate today's newsletter?Vote below to help us improve the newsletter for you. |
If you have specific feedback or anything interesting you’d like to share, please let us know by replying to this email.
Join the conversation