- The Daily Update
- Posts
- Nvidia unveils AI superchip
Nvidia unveils AI superchip
Research: Medical AI finds Alzheimer's genes
Jack Shields
November 14, 2023
Welcome back!
Today’s research spotlight showcases another exciting possibility for medical AI. We’ll also take a look at a Nvidia’s new AI chip and OpenAI’s efforts to attract the industry’s top researchers.
Let’s dive in.
In today’s Daily Update:
🗞️ Nvidia unveils new top-of-the-line AI chip
🤖 AI shows promise for finding Alzheimer’s genes
📸 OpenAI recruits Google AI talent with multimillion-dollar offers
🚨 AI Roundup: Three quick hits
Read time: 2 minutes
LATEST NEWS
🗞️ Nvidia unveils new top-of-the-line AI chip
Source: Nvidia
Nvidia just unveiled the HGX H200 chip, a new GPU that is more capable of handling intensive generative AI work.
Key points:
The new GPU upgrades Nvidia’s H100 with 1.4x more memory bandwidth and 1.8x more memory capacity.
Nvidia will release the first H200 chips in the second quarter of 2024.
Apart from memory upgrades, the H200 appears to be substantially the same as the H100.
Nvidia has already been overwhelmed by the demand for its H100 chips. The company did not publicly outline a plan for H200 production.
What this means: Faster and more extensive memory will improve the GPU’s performance across computationally demanding tasks like generative AI models. As the leader in the AI chip market, Nvidia’s upgraded H200 chip will usher in the next wave of advanced AI models.
RESEARCH INSIGHT
🤖 AI shows promise for finding Alzheimer’s genes
Source: Adobe Firefly
Researchers across the U.S. are testing AI-based approaches to identify Alzheimer’s genes. Initial results are promising.
What you should know:
One AI algorithm sorts through large numbers of brain images and identifies scans that include characteristics of Alzheimer’s.
Another machine-learning method identifies important structural features of the brain that could help scientists spot new signs of Alzheimer’s in brain scans.
The AI classifier reportedly detected Alzheimer’s in brain scans with over 90% accuracy.
Efforts to develop these algorithms are led by Paul Thompson, a neuroscientist at the University of Southern California in Los Angeles.
The bottom line: Medical AI is still advancing rapidly. Advanced algorithms will soon unlock new treatment options for a plethora of diseases and health conditions.
BUSINESS SPOTLIGHT
📸 OpenAI recruits Google AI talent with multimillion-dollar offers
Source: OpenAI
OpenAI is enticing senior AI researchers at Google to join them by offering compensation packages worth up to $10 million per year.
The details:
OpenAI is specifically targeting employees working on Google’s Gemini family of AI models.
These offers come as OpenAI nears the completion of an employee stock sale that could triple its valuation to more than $80 billion.
The company expects to have a computing advantage over Google sometime next year after Microsoft launches its own AI chips.
Why it matters: The race to secure top AI worker talent is just as intense as the race to develop top AI models. Now is a great time to be an AI researcher.
MORE TRENDING NEWS
🚨 AI Roundup: Three quick hits
Source: Adobe Firefly
New Rules: YouTube will start cracking down on AI clones of musicians.
Smart Notes: Notion’s new Q&A feature lets users ask AI for details from their notes.
Regulation Talks: Negotiations about Europe’s AI Act have reportedly reached a stalemate.
THAT’S ALL FOR TODAY