- The AI Daily Brief
- Posts
- Is Mysterious “GPT2” Actually GPT-5?
Is Mysterious “GPT2” Actually GPT-5?
Plus Apple Poaching Googler’s for Secret AI Lab
The AI Breakdown First Five - Tuesday April 30, 2024
Today on the First Five:
5. Microsoft Invests $1.7B in Indonesia
4. Biden Admin Announces New Round of AI Actions
3. Apple Poaching Googler’s for Secret AI Lab
2. Another Controversial AI Device Review
1. Is Mysterious “GPT2” Actually GPT-5?
LAST CHANCE: Join NLW’s May AI Learning Cohort
Superintelligent is the platform for useful AI learning with 300+ video tutorials. This is your last chance to sign up for my personally hosted cohort that will have its own Discord channel, tutorial recommendations and more.
Sign up on https://besuper.ai/ with code nlwmay. You’ll get 25% off your first month and also be automatically added to the cohort!
5. Microsoft Invests $1.7B in Indonesia
Microsoft’s global tour for AI and cloud investment continues. CEO Satya Nadella announced that the company will be spending $1.7B over the next four years, and alongside the commitment will help train 2.5m people in the region with AI skills.
Today we announced our largest investment in our nearly three decades in Indonesia, as we partner to advance the country’s cloud and AI ambitions, and help Indonesia thrive in this new era.
— Satya Nadella (@satyanadella)
4:30 AM • Apr 30, 2024
4. Biden Admin Announces New Round of AI Actions
When the Biden Admin launched their AI executive order last year, agencies had a set of commitments at intervals including 30, 90, 120 and 180 days. The 180 day mark has just passed and the White House has announced dozens of updates.
3. Apple Poaching Googler’s for Secret AI Lab
The FT reports that a secret Zurich-based lab connected with Apple’s AI efforts has been poaching talent from around the industry, with a particular focus on Google, from whom it has poached 36 leaders over the last few years.
Apple Hired Dozens of AI Experts From Google for a Secretive Zurich Research Lab
— MacRumors.com (@MacRumors)
8:20 AM • Apr 30, 2024
2. Another Controversial AI Device Review
A week after his controversial Humane Pin review, Marques Brownlee has called the R1 “Barely Reviewable.” While entrepreneurs flocked to critique him for not seeing the vision, Brownlee argued that the release is part of a broader trend of releasing crappy products before they’re done. What do you think?
NEW VIDEO - Rabbit R1: Barely Reviewable
youtu.be/ddTV12hErTc
This is the pinnacle of a trend that's been annoying for years: Delivering barely finished products to win a "race" and then continuing to build them after charging full price. Games, phones, cars, now AI in a box
— Marques Brownlee (@MKBHD)
12:21 AM • Apr 30, 2024
1. Is Mysterious “GPT2” Actually GPT-5?
When a model called “gpt2-chatbot” appeared on an LLM benchmarking site, people quickly noticed that it seemed, well, a lot better than GPT-2. Soon speculation began that it was actually a version of GPT-4.5 or even GPT-5 out in the wild.
There is a mysterious new model called gpt2-chatbot accessible from a major LLM benchmarking site. No one knows who made it or what it is, but I have been playing with it a little and it appears to be in the same rough ability level as GPT-4. A mysterious GPT-4 class model? Neat!
— Ethan Mollick (@emollick)
4:57 PM • Apr 29, 2024
i do have a soft spot for gpt2
— Sam Altman (@sama)
12:44 AM • Apr 30, 2024
BONUS: China’s “Sora Killer?” or…not so much?
China's first Sora-level model Vidu released: it can generate up to 16 seconds and up to 1080P video.
Sora Killer or Temu Sora? What do you think?
— Theoretically Media (@TheoMediaAI)
2:28 PM • Apr 27, 2024
The Latest Episodes of The AI Breakdown
Thanks for reading! How did you find today’s edition?
How was today's newsletter? |
The AI Breakdown podcast - https://pod.link/1680633614
The AI Breakdown YouTube - https://www.youtube.com/@theaibreakdown
The AI Breakdown Discord - https://bit.ly/aibreakdown