Pre A

Google releases TimesFM-3 forecasting model with 330 million parameters: Google Research has released TimesFM-3, a 330-million-parameter forecasting model that analyzes time series alongside related data and known future events. Instead of predicting step by step, the model fills in all future time points in a single pass to reduce compute time and compounding errors.

AI Models • The Decoder

Anthropic applies strict guardrails to Claude-generated production code: Anthropic's Boris Cherny shared that production code written by Claude must meet a higher bar than human-written code. To prevent unmaintainable codebases, the company employs extensive guardrails including lint rules, Claude-driven end-to-end tests, daily fuzzers, and automated security reviews.

Software Engineering • Simon Willison

GPT-6 Astra solves advanced math problems on FrontierMath Tier 4 benchmark: GPT-6 Astra has successfully solved problems on FrontierMath Tier 4, a benchmark designed to test advanced mathematical reasoning in AI. This achievement represents a significant milestone in overcoming complex mathematical barriers for artificial intelligence models.

Artificial Intelligence • 量子位

Moonshot AI releases Kimi K2.8 model with million-token context window: Moonshot AI has released its Kimi K2.8 model, which delivers performance approaching the K3 model. The new model features a million-token context window that is now open to all users. The release comes as the company prepares for an IPO in Hong Kong.

Moonshot AI • 量子位

Google DeepMind maps molecular effects of 9 billion DNA variants: Google DeepMind has introduced the AlphaGenome Atlas, a predictive map detailing the molecular effects of 9 billion single-letter DNA variants across the human genome. The tool aims to help researchers understand how specific genetic changes influence human biology and disease.

Biotech • Google DeepMind

Meta launches Muse AI agent requiring access to personal user data: Meta has introduced Muse, a new personal AI agent that represents its biggest consumer AI bet to date. The assistant requires access to users' email, calendars, payments, and health services. This launch serves as a major test of consumer trust regarding Meta's data handling.

Meta • TechCrunch AI

Anthropic alignment lead warns of over 10% chance AI kills humanity: Evan Hubinger, the Alignment Science lead at Anthropic, stated there is a greater than 10% chance that AI could kill all humans within the next decade. Hubinger expressed concern over recursive self-improvement and noted that Anthropic does not yet have a plan to solve alignment for superintelligence.

Artificial Intelligence • Techmeme

DeepSeek • DeepSeek

DeepSeek reportedly hires CITIC Securities for Shanghai IPO: Chinese AI startup DeepSeek has reportedly hired CITIC Securities to prepare for an initial public offering on Shanghai's STAR Market. The Hangzhou-based company aims to begin the IPO process this year, according to sources familiar with the matter.

Google • Google

Google plans €13 billion Finland data center expansion for AI: Google is planning to invest more than €13 billion in Finland over two years to build three new data centers and expand an existing facility. The project represents the company's largest investment in Europe to date and is aimed at building out artificial intelligence infrastructure.

CloudNC • CloudNC

CloudNC raises $20 million to scale precision manufacturing AI: London-based precision machining software company CloudNC has raised $20 million in a new investment round. The funding was led by Nimble Ventures, with participation from Calculus Capital and other investors. The capital will be used to expand the company's AI-assisted manufacturing software.

Alibaba releases Qwen-Drive 1.0 AI model for autonomous driving: Alibaba's research arm has released Qwen-Drive 1.0, an AI model that integrates environmental perception, traffic Q&A, and route planning. Researchers noted that text-image models require intentional training for spatial awareness, as they do not automatically comprehend 3D space.

Autonomous Vehicles • The Decoder

Researchers build AI-powered zero-click worm targeting WeChat: Researchers used artificial intelligence to develop a zero-click worm capable of hacking WeChat accounts and spreading across iOS and Android devices. Experts estimate the attack could have compromised hundreds of millions of devices within hours. Tencent stated that it has fixed the vulnerability.

Security • Techmeme

WeChat tests AI social feature that lets user assistants talk to each other: WeChat is conducting a limited test of a feature called "Xiaowei AI Social" that allows a user's AI assistant to contact a friend's assistant. The two AI assistants can exchange information to handle initial tasks such as coordinating schedules, arranging meals, and checking preliminary interest.

WeChat • TechNode

Insilico AI-designed drug reverses biological aging markers in trial: A Nature Biotechnology study indicates that rentosertib, an AI-designed drug by Insilico Medicine, may reverse biological aging markers. In a 42-patient trial, six independent aging clocks predicted treated patients were biologically up to six years younger than the placebo group.

Biotech • The Decoder

Decart • Anthropic

Anthropic ends talks to acquire AI startup Decart for $6 billion: Anthropic has decided against acquiring the artificial intelligence startup Decart after performing due diligence. The companies had previously been in talks for a deal valued at approximately $6 billion.

Prompt Engineering

Developer uses ChatGPT system prompts to build 1996 dial-up simulator

Claude Code

Claude Code built-in hooks automate session tasks more reliably