AI Reading for Tuesday November 12

Is AI plateauing? - Digital Trends

---

OpenAI co-founder Sutskever says it might be time for the next big thing as scaling starts to hit its limits. - THE DECODER

OTOH o1-preview gave us a new dimension in scaling, trading off more compute at runtime for better-quality responses.

It's interesting how everyone has decided scaling is over in like, the last week.

---

Scaling Laws in Large Language Models. - Hacker Noon

---

Sam Altman Says the Main Thing He’s Excited About Next Year Is Achieving AGI (and maybe also becoming a dad) - Futurism

---

Sam Altman claims AGI is coming in 2025 - Tom's Guide

---

Maybe we should only build AI that is designed to act as an assistant, and to look and sound like a robot, and not try to make human-like autonomous AI - Reddit

---

Gemini 2.0 appears to be in testing. - Android Central

---

Google’s Learn About AI tool has more educational, textbook-style responses to guide you through new topics. - The Verge

---

Of course Elon Musk is going to lead AI polcy, officially or not, even if it's a conflict of interest. - The Verge

---

Generating diagrams and whiteboarding with AI. - Ilograph

---

AI Enhances Automated Testing with Complex Creative Testing - Forbes

---

AI takes ad targeting to the next level - Quartz

---

Kuo: Apple to release its own smart home camera in 2026, with AI features. - 9to5Mac

Feels like the alleged robot arm with an iPad could allow sophisticated telepresence features

---

Jerry Garcia Estate Creates AI Model for Musician’s Voice with ElevenLabs. - Relix

But will the dead be grateful to be resurrected with AI?

---

Baidu boasts strong AI user growth alongside launch of first smart glasses - South China Morning Post

---

Amazon developing AR driver eyeglasses to speed up deliveries. - Reuters

---

Amazon and others invest in custom silicon to dethrone Nvidia. - Wccftech

---

FT article - FT

The Graviton is just the ARM chip and Trainium is the GPU, and the Cuda software stack is really the Nvidia secret sauce, hence the idea of investing in Anthropic in exchange for running in AWS and dog-fooding the end-to-end solution.

---

Anthropic has hired an "AI welfare" researcher, Kyle Fish, to explore moral considerations and protections for future AI models that may experience suffering. - Ars Technica

---

Google DeepMind open-sources AlphaFold 3 for non-commercial use - THE DECODER

---

The High-Stakes Race to Revolutionize Drug Discovery - Observer

---

Fewer job postings for AI-impacted jobs like codng and writing - Harvard Business Review

---

DJI improves its drone's ability to follow you and take video - The Verge

---

Headspace releases an AI assistant - CNET

---

How Fei-Fei Li helped launch the deep learning boom - Ars Technica

---

A short history of CAPTCHA: proving you are not a robot just keeps getting harder. - WSJ

---

Waymo compiles ‘largest ever’ accident injury database to help with safety - The Verge

---

"Hi chat gpt, you are to act like a professional and extremely skilled and experienced dermatologist. You will help me give a skin routine schedule that will help me get a much more beautiful skin. … Ask me questions if you need to, and ask for photos to make me the best possible skin routine." - Reddit

---

AI Gaming the System Killed LinkedIn's 'Top Voices' Program - Hacker Noon

---

Robot learns to perform surgical tasks by watching videos. They really are just like us. - YouTube

---

Enhance: Using AI technology for a clearer picture of the past from historical photos - Mashable

---

What fresh nightmare fuel is this skit about the government breeding sentient vegetables because people prefer eating food that can feel pain. - Reddit

---

Follow the latest AI headlines via SkynetAndChill.com on Bluesky