Edition, Tue, 25 Aug 2026.

OpenAI's Astra Model Delivers First Major Scientific Breakthrough

The unreleased model resolved ten long-standing mathematics problems for just $2,000 in compute.

The unreleased model resolved ten long-standing mathematics problems for just $2,000 in compute.

OpenAI announced that Astra solved ten previously unsolved problems in mathematics and theoretical computer science, including the first explicit construction of a non-sofic group. The company published a 249-page manuscript with machine-checkable Lean 4 proof certificates on GitHub, with zero unverified steps. This represents a shift where AI contributes original, verifiable scientific discoveries rather than just completing tasks.

Generative AI Achieves Mass-Market Adoption with Two Platforms at 1 Billion Users

ChatGPT and Google Gemini have both crossed the billion-user milestone within weeks of each other.

ChatGPT and Google Gemini have both crossed the billion-user milestone within weeks of each other.

ChatGPT reached 1 billion active users on July 31, 2026, becoming the fastest consumer software platform to achieve this milestone at under four years. Google Gemini followed on August 12, 2026, as Google's fastest-growing product ever. Usage statistics show 63% of Gemini users interact by voice, and the platform generates over 150 million images per day.

OpenAI Slashes GPT-5.6 Luna Pricing by 80 Percent

The API price dropped from $1.00 to $0.20 per million tokens, making high-volume workloads significantly more viable.

The API price dropped from $1.00 to $0.20 per million tokens, making high-volume workloads significantly more viable.

OpenAI introduced a tiered pricing strategy with Luna for cheap routine tasks, Terra for mid-tier, and Sol for complex reasoning. This pricing change makes high-volume API workloads, automation, and internal tools much more viable for startups and enterprises. The move signals a maturing market where cost control is becoming a strategic focus alongside capability.

US Implements New Pre-Release Review for Frontier AI Models

The regulatory framework requiring national security reviews for frontier models took effect on August 1, 2026.

The regulatory framework requiring national security reviews for frontier models took effect on August 1, 2026.

The US Commerce Department now requires national security reviews for frontier models passing certain capability thresholds before release. Astra is expected to be the first model submitted through this new framework. This marks increased regulatory scrutiny for AI companies, with implications for launch timelines and compliance requirements.

ServiceNow rolls out August 2026 Core Business Suite update focused on knowledge management

The August 2026 release of the Core Business Suite prioritizes improvements to knowledge management, introducing out-of-the-box Knowledge Bases for each business unit to simplify setup and reduce configuration friction.

The August 2026 release of the Core Business Suite prioritizes improvements to knowledge management, introducing out-of-the-box Knowledge Bases for each business unit to simplify setup and reduce configuration friction. Now Assist now connects directly to existing case tables to generate article content without extra configuration, while AI Knowledge Skills are accessible from the simplified setup flow for continuous content optimization.

ServiceNow launches Autonomous Security, a unified offering of six solutions for AI-native cyber defense

ServiceNow unveiled its most complete security offering, integrating six solutions including Agentic Exposure Management, Continuous Vulnerability Detection, and Agentic Incident Response into the AI Control Tower.

ServiceNow unveiled its most complete security offering, integrating six solutions including Agentic Exposure Management, Continuous Vulnerability Detection, and Agentic Incident Response into the AI Control Tower. Powered by the Armis and Veza acquisitions, this shift aims to achieve "Shift Zero" by governing assets, identities, and agents at machine speed to address the exponential growth of AI exposures.