Peter Thum

This is hilarious. This is what AI was made for. I love it. 100% accurate.

Humanoid robots are the "next big thing" and are already widely used in industry.

Jensen just addressed DLSS5 again at a press conference because the question arose as to why there are discussions about it in the community. To be fair, his statement sounds a little different from what I've heard from other NVIDIA employees. „It adds generative AI but doesn’t change artist direction.“ To be fair, this does indeed sound different than what I heared before. (Sorry for the bad sound)

Valley Bets on Chinese AI! That's a surprise, but it makes sense when you think about value for money: NBC’s reporting shows a growing chunk of Silicon Valley quietly building on free, open-source Chinese models like DeepSeek and Alibaba’s Qwen instead of pricey closed systems from OpenAI and Anthropic. Founders say these Chinese models are now “surprisingly close to the frontier,” cheap to fine-tune, and easy to self-host, which makes them attractive for cost-sensitive products and privacy-critical workloads.

1/ Tell me again how blue-collar workers wont be replaced in the very near future. The robot Figure 02 had an 11-month deployment at BMW Group’s Spartanburg plant, including 10-hour daily shifts, during which it moved 90,000+ parts, ran 1,250+ hours, and contributed to the production of over 30,000 X3 vehicles. The deployment yielded critical learnings (especially about reliability and mechanical design) that are being incorporated into the next-gen robot Figure 03.

Satya Nadella is playing 4D-Chess at this pint Anthropic is partnering with Microsoft and NVIDIA: Microsoft will host Claude on Azure and NVIDIA’s architecture will power it. Anthropic has committed to purchasing US$30 billion of Azure compute capacity and potentially up to one gigawatt of compute capacity using NVIDIA’s systems to support its growth.

Amazing update: Scientists Have Created A Robot Eye With Better Sight Than Humans! Scientists at Georgia Tech have built a new robotic “eye” using a photoresponsive hydrogel lens that can focus without electricity. The lens can detect incredibly fine details - like the hair on an ant’s leg - thanks to light-driven swelling and shrinking of the hydrogel. This tech could transform soft robotics, medical tools, wearable devices, and even microscopes with its high precision and autonomous light-powered operation.

A few thoughts on the Kimi-K2-thinking release 1) Open source is catching up and in some cases surpassing closed source. 2) More surprising: the rise of open source isn’t coming from the U.S., as was still expected at the end of last year (LLama), but from China—kicked off in early 2025 with DeepSeek, and now confirmed by Kimi. 3)The sanctions against China aren’t achieving their intended effect; on the contrary, the chip restrictions are pushing Chinese scientists to become far more creative, essentially turning necessity into a virtue. 4)Taken together, this leads to the conclusion that China is not just becoming an increasingly strong competitor to the U.S., but is now catching up—and it’s no longer clear who will win the race toward AGI.

Today's newsletter features something very special: another exclusive interview, this time with Nadia Harhen, General Manager of AI Simulation at SandboxAQ, a Google spin-off. We talked about LQMs (Large Quantitative Models) and the future of AI in research. It was a very exciting conversation! Enjoy reading, link below!