- The Daily Update
- Posts
- Biden backs AI
Welcome back!
More big news from Washington today as the U.S. aims to improve its competitiveness in the technology sector.
Nvidia also just announced a huge breakthrough in robotics and we have Apple’s AI master plan on deck in today’s business spotlight.
In today’s Daily Update:
🗞️ White House designates 31 tech hubs to focus on AI
🤖 Nvidia enables robots to teach themselves new skills
📸 Apple plans to integrate AI across all devices
🚨 AI Roundup: Four quick hits
Read time: 2 minutes
LATEST NEWS
🗞️ White House designates 31 tech hubs to focus on AI
Source: Adobe Firefly
Yesterday the White House announced it is designating 31 technology hubs to improve American competitiveness in the technology sector.
What you need to know:
The hubs are located in 32 states and Puerto Rico. They will be able to compete for $40 million to $75 million each in grants.
“For too long, economic growth and opportunity has clustered in a few cities on the coasts,” the White House said in a press release.
The move was authorized under the CHIPS and Science Act, which President Joe Biden signed in August 2022. The act authorized $10 billion to invest in technology hubs nationwide.
Hubs will focus on a wide range of technological areas, including quantum computing, artificial intelligence, clean energy, medicine and biotechnology.
Crucial quote: A tech hub designation is “a strong endorsement of a region’s plan to supercharge a critical technology ecosystem and become a global leader over the next decade,” the U.S. Economic Development Administration said.
RESEARCH INSIGHT
🤖 Nvidia enables robots to teach themselves new skills
Nvidia researchers just developed an AI agent called Eureka that can teach robots complex skills as well as a human can.
Key points:
Eureka can teach robots nearly 30 tasks, including how to open drawers and cabinets, toss and catch balls, and use scissors.
Developers can use NVIDIA Isaac Gym to experiment with the project’s algorithms. Eureka itself is powered by OpenAI’s GPT-4 large language model.
According to the research, Eureka-generated reward programs (which enable trial-and-error learning) outperform expert human-written ones on more than 80% of tasks.
This leads to an average performance improvement of more than 50% for the bots.
Why it matters: Eureka is pushing robots to learn new skills as dynamically as humans (see for yourself in the video above). AI agents that generate algorithms to train robots could supercharge advancements in robotics.
BUSINESS SPOTLIGHT
📸 Apple plans to integrate AI across all devices
Source: Adobe Firefly
Apple is reportedly working to build generative AI into all of its products in an effort to catch up to competitors according to Bloomberg.
The details:
Apple is building a revamped version of Siri powered by its proprietary large language model, which will be deeply integrated into its products.
The company also plans to roll out smarter auto-complete in Messages, auto-generated playlists in Music, code suggestions in Xcode and more.
There are internal debates about how to deploy AI: entirely as an on-device experience, a cloud-based setup or something in between.
Concerns linger about whether Apple’s AI tech can match the strength of models like ChatGPT and Claude.
The bigger picture: Bloomberg reports that Apple executives were caught off guard by the AI wave, and have since been scrambling to catch up to rivals. Whether this is true or not, Apple has been noticeably quiet as its competitors race to launch consumer-facing AI models. Expect big AI announcements from Apple soon.
MORE TRENDING NEWS
🚨 AI Roundup: Four quick hits
Source: Adobe Firefly
Pay Up: Major newspapers want payments for articles used to train ChatGPT.
Future-proof: IBM CEO Arvind Krishna says students should develop critical thinking skills and study foundational subjects like history to future-proof themselves for the AI economy.
3D-GPT: Researchers unveil new AI system that can generate 3D worlds from simple text descriptions.
Fact Check: New study shows AI can improve Wikipedia’s accuracy by suggesting better references in nearly 50% of cases.
THAT’S ALL FOR TODAY