AI
OpenAI's GPT-6 Astra designed a working hardware device in under a week
OpenAI says GPT-6 Astra handled enclosure, PCB layout, firmware and UI for a Codex device in under a week, while Anthropic filed for a record IPO.
This edition was produced with artificial intelligence. Text and voice are generated automatically.
OpenAI says GPT-6 Astra designed a hardware device end to end in under a week
OpenAI’s hardware team used GPT-6 Astra to design a device for talking to the Codex coding agent, completed as a side project in under a week. According to OpenAI, Astra handled enclosure design, PCB layout with component selection and trace routing, firmware and the user interface. A contract manufacturer built a small batch of prototypes in under a month, though several revision rounds were needed because Astra did not always account for manufacturing constraints, tolerances and assembly details. A newer internal model scores 0.875 on the BenchCAD 2.0 benchmark, while GPT-6 Astra at its highest reasoning level reaches 0.594. OpenAI says token usage across its hardware teams has grown 23x since February.
OpenAI says GPT-6 Astra designed a hardware device from scratch in under a week →
Topics: GPT-6 AstraOpenAI
Anthropic’s Message Threads beta cuts time to first token by about 35 percent
Anthropic released Message Threads as a public beta on its Claude platform. Previously every API request had to resend the entire conversation history; Message Threads transmits only what has changed since the last message. Anthropic says this cuts time to first token by about 35 percent for requests between 500,000 and one million tokens. Coding agents, computer-use agents and workflows processing large volumes of documents or images benefit most. The thread stays encrypted in memory between requests and expires after five minutes of inactivity. Token usage does not change, since Claude still reads the full conversation history. The feature works on the Claude platform and through AWS, Microsoft Foundry and Google Cloud.
Anthropic’s Message Threads cut time to first token by 35 percent for long conversations →
Stanford study finds 44 percent of cited LLM claims have already expired
Stanford researchers Owen Queen and James Zou tested 50 widely cited claims about large language models against 20 models and found 44 percent have already expired. Among the 22 findings with an estimated expiration date, the average lifespan was 15.7 months. Another 34 percent weakened significantly, only 16 percent still hold and 6 percent have gotten stronger. Claims about what models cannot do expire fastest: five of nine reasoning-related findings are expired, including counting letters in strawberry and arithmetic in unfamiliar number systems. In a case study on fabricated citations, GPT-3.5 made up 55 percent of references in 2023, while GPT-6.1 Sol fabricated 0 percent and Claude Sonnet 5.5 came in at 1 percent.
Rapid LLM development is outrunning the research meant to explain it, study finds →
AI agents can now steer SETI’s Allen Telescope Array and analyze signals live
At the Allen Telescope Array in Northern California, AI agents can now steer the telescope and analyze incoming radio signals on the fly, though researchers still type in their commands for now. The system runs on Nvidia’s Holoscan platform, which the SETI Institute and Breakthrough Listen use to process telescope data in real time. Nvidia says the hardware can process up to 2.4 terabits per second. Work that used to take weeks can now wrap up in hours, according to the team, which says it detected a pulsar in the Crab Nebula in real time using AI, a first. The team now wants to test open-source software that can run locally on the edge, with an eye toward future sensors on the far side of the Moon.
Topics: Nvidia
Anthropic S-1 shows $1.7 billion quarterly personnel bill across 5,300 staff
Anthropic’s amended S-1 filing shows more than $1.7 billion in personnel expenses from July through September, with 5,300 employees, working out to over $322,000 per person over the three-month span, or about $107,300 per employee per month. Those costs include cash salaries along with bonuses, benefits and employer payroll taxes. The largest share, $1.1 billion, came from stock-based compensation. In the first nine months of 2026, personnel costs reached $4.8 billion, or about 15 percent of the $33.2 billion in revenue reported for the same period. Anthropic has filed for what could be the biggest IPO ever and has grown its workforce fivefold since the end of 2024.
Topics: Anthropic