OpenAI Cuts GPT-5.6 Sol Prices Amid Copilot Breach

OpenAI has reduced the pricing of its flagship GPT-5.6 Sol model by 50%, bringing costs down to $2.50 per million input tokens and $15 per million output tokens. The model also features a massive 1.05 million token context window and improved system uptime across three providers.

DuckDB v2.0 is scheduled for release this fall, introducing major architectural upgrades to the popular database engine. Headline features include the ability to run DuckDB as a server and support for asynchronous I/O. The update also adds new capabilities like triggers, a VARIANT data type, a revamped SQL parser, and a completely new storage format.

A developer has successfully implemented a working driver for the Fairphone 6's main camera on PostmarketOS, building upon previous efforts for the wide-angle lens. The new driver supports autofocus and color correction, although the color tuning remains a work in progress.

Researchers have identified the first confirmed case of a death caused by a medieval trebuchet after examining a skeleton from the fourteenth century. The remains exhibit catastrophic trauma consistent with being struck by a massive boulder launched from the siege engine. This discovery provides rare physical evidence of the lethal impact of early artillery on medieval warriors.

Researchers have published a new paper detailing a Rust-based framework for GPU offloading that promises high computational performance alongside memory safety and cross-platform portability. This development underscores Rust's expanding role in high-performance computing and hardware acceleration.

Wiz's Red Agent successfully compromised Snowflake's internal Jira platform by exploiting a security flaw in a pull request assisted by GitHub Copilot. The incident highlights the potential vulnerabilities introduced when developers rely on AI-generated code autofixes without adequate manual oversight.

Readers are discussing the perception of chimerical or impossible colors, which can be experienced through specific optical illusions that temporarily saturate the eye's color cones. The thread highlights several interactive demonstrations and a new visualization tool that lets users explore the gamut of these imaginary colors to generate unique contrast-sensitive palettes for web and software design.

This article explores the development of MS-DOS 2.0, highlighting its major architectural shift toward a UNIX-inspired hierarchical file system. The update moved the software beyond its original CP/M roots to support larger hard drives and subdirectories. Ultimately, this release laid the foundational directory structure that would influence future PC operating systems for decades.

Even strong proponents of artificial intelligence are growing frustrated with the overwhelming amount of low-quality AI-generated text flooding the internet. In response, users are adopting a new acronym called AI;DR, a play on the traditional TL;DR, to quickly dismiss and ignore these automated walls of slop.

OpenAI has released GPT-5.6 Sol, its most powerful vision model to date, outperforming previous iterations like Terra and Luna in tasks such as OCR, counting, and extraction. Comprehensive testing shows that Sol delivers superior accuracy while maintaining highly competitive speed and cost efficiency when benchmarked against other leading vision language models.

The author shares a follow-up on their highly discussed transition from Gmail to Fastmail after several months of using the alternative service. This update details their ongoing experience and the practical realities of making the switch.

Dutch photographer Pie Aerts has captured the isolated lives of the puesteros, solitary ranch workers in Chilean Patagonia who tend to livestock for minimal pay. Writer Ana Karina Zatarain highlights these striking images, which showcase the extreme solitude of these men who often go months without human interaction.

India has paved the way to introduce merchant fees for UPI transactions, marking a significant shift in its digital payments ecosystem. This development raises critical questions about who will ultimately bear the costs and whether consumers will resist the new charges.

This technical overview explores the mechanics of stack memory allocation, detailing how functions like alloca dynamically adjust the stack pointer. It also covers the concept of stack probing, explaining how operating systems use guard pages to safely handle memory access and prevent segmentation faults.

Israel has created a fake think tank named the Hanover Institute in an apparent attempt to manipulate AI chatbots. Within just a week, the organization published over 100 articles specifically tailored to influence the responses of these language models. This operation aims to shape AI-generated information by flooding the internet with targeted content.

YC S26 startup Speko has launched as an OpenRouter for Voice AI that dynamically routes user sessions to the best-performing models. The platform addresses the industry's reliance on English-centric benchmarks by evaluating 61 speech and language models across 10 different languages. This ensures that non-English speakers consistently receive the highest quality voice AI experience.

California has implemented the nation's first tire efficiency standards to address an often overlooked factor in overall fuel economy. These new regulations are projected to save drivers approximately one billion dollars annually at the gas pump.

A recent GitHub service degradation is causing disruptions for Cursor Origin, the newly launched Git platform from Cursor. This ongoing outage is currently affecting several key Cursor services, specifically Automations, Cloud Agents, and Codebase. Users are advised to monitor the Cursor status page for real-time updates on the resolution.

A recent mathematical digest provides a detailed breakdown of the proof for Sendov's conjecture, a complex problem regarding the roots of polynomials and their derivatives, alongside a strengthened variation proposed by Phelps and Rodriguez. This overview offers valuable insights for readers interested in theoretical mathematics and its foundational applications in computer science.

A newly proposed technique addresses large language model tail latency by simply sending every request twice and returning the faster response. This redundant execution strategy effectively bypasses slow outlier requests to significantly improve overall system responsiveness. Ultimately, it provides developers with a highly practical and low-complexity solution for optimizing LLM performance.