TECH

OpenAI's new science blog: a physicist used GPT-6 Astra to help build a candidate lattice definition of the Standard Model

Shadow News
@ShadowNewsHQ
2:01 AM ยท Nov 17, 2026
View on X

๐Ÿ’ป TECH

GPT-6 Astra just helped tackle a deep physics puzzle โš›๏ธ

In the first post on OpenAI's new science blog, UC Santa Barbara physicist Nathaniel Craig says Astra took the lead on the detailed proof for a candidate lattice definition of the Standard Model. Outside experts

Quoting OpenAI (@OpenAI)

Physicist Nathaniel Craig shares how he worked with GPT-6 Astra to develop a candidate lattice definition of the Standard Model. Itโ€™s the first post on our new science blog:

OpenAI launched a new science blog, and its first post is a big one for physics. In a guest post, UC Santa Barbara physicist Nathaniel Craig describes work on a candidate nonperturbative lattice definition of the Standard Model, including its chiral fermions. Craig says GPT-6 Astra took the lead on the detailed proof. The work, first shared in an October preprint, was reviewed by outside experts Maarten Golterman, Andreas Kronfeld and Yigal Shamir. Craig calls the result "more a beginning than an end."

TECH

Judge lets DRAM supply-conspiracy lawsuit against Samsung, SK hynix and Micron move forward

A federal judge in California has let a DRAM lawsuit against Samsung, SK hynix and Micron move forward. On November 14, Judge Noรซl Wise of the Northern District of California denied the companies' joint motion to dismiss Garciaguirre v. Samsung. The suit alleges the three makers, which control more than 91% of DRAM revenue, cut regular memory output from late 2022 while shifting to high-bandwidth memory for AI, and claims prices rose 697% over 18 months. The judge wrote that the companies' moves looked like "hawks flying in formation." The allegations are unproven. The companies must answer by December 7, with a case management conference on January 12, 2027.

๐Ÿ’ฌ Comment
Share
TECH

Gemini 4 Hydrogen is now Google's everyday AI model, replacing 3.8 Flash for everyone

Google's Gemini 4 Hydrogen is now the everyday Gemini model. Announced November 12 on Google's The Keyword blog, it replaces Gemini 3.8 Flash across Google Search, the Gemini app (where free users get it through automatic model selection), Google AI Studio, Android Studio, Gemini for business and the Gemini API. Google DeepMind says Hydrogen finishes tasks 2.5 times faster than 3.8 Flash, and it scores 51 on the Artificial Analysis Intelligence Index, just ahead of the larger Gemini 4 Argon at 50. API pricing is $1.50 per million input tokens and $7.50 per million output tokens, flat across its 1 million token context window.

๐Ÿ’ฌ Comment
Share
TECH

Report: Samsung plans a $1.9 billion Austin fab expansion to make SpaceX Starlink chips

Report: Samsung plans a $1.9 billion expansion of its Austin, Texas chip plant to make chips used in SpaceX's Starlink satellites, according to Seoul Economic Daily. The expansion would add 18nm FD-SOI capacity to build STMicroelectronics' STM32V8 microcontroller, which is used in Starlink's satellite laser links, with production targeted for 2030. STMicroelectronics calls the STM32V8 the industry's first 18nm microcontroller and developed the process with Samsung Foundry. Samsung has not confirmed the report.

๐Ÿ’ฌ Comment
Share

Follow @ShadowNewsHQ

Breaking news, official trailers and the daily 9 PM ET recap, straight in your feed on X.

Follow on X

Like this page? Follow us on X and share it with a friend.

Follow @ShadowNewsHQ