Legacy Code

Daily AI intelligence for business professionals

LLMs & Models

New model releases, benchmarks, and breakthroughs — explained in plain English.

OpenAI Launches Financial Services Edition of ChatGPT with Market Data Integration

OpenAI Releases GPT-6 Astra, Its Most Capable Business AI Model Yet

OpenAI's AI Agents Solve Navier-Stokes Millennium Prize Problem

GPT-5.6 Sol Enables Autonomous Quantum Computing Research at Scale

OpenAI Claims Mathematical Breakthrough on Navier-Stokes Problem, Drawing Academic Scrutiny

Anthropic's Small Team Emerges as Consistent AI Research Producer

Claude AI Formalizes Fermat's Last Theorem in 11 Days—A Milestone in Mathematical AI

Anthropic's Claude Gains Ability to Run Science Experiments Autonomously

OpenAI's GPT-6 Astra Reaches Critical Cybersecurity Capability; Early Users Report Major Efficiency Gains

OpenAI Releases GPT-6 Astra, Claims Entry Into AGI Era With Advanced Computer Use

Anthropic Releases Claude Fable and Mythos Models with 75% Cost Reduction

ChatGPT Now Integrates with Healthcare Systems and Patient Records

OpenAI's Astra Model Introduces Critical Cybersecurity Risks—and Safeguards

OpenAI's ChatGPT Advertising Tier Reaches $1 Billion Annual Revenue Milestone

Polimill Leverages OpenAI Models to Modernize Japan's Municipal Government Infrastructure

Anthropic Shows Progress on Self-Improving AI Systems

Open-Weight AI Models Become Hottest M&A Targets in Silicon Valley

OpenAI Ends Model Access to Cursor After SpaceX Acquisition

The Full Stack Behind Abundant Intelligence: OpenAI's Cost and Performance Vision

OpenAI's Jalapeño Chip Delivers Faster, More Efficient AI Processing

Anthropic Integrates AI-Generated Watermarks Into Claude Output

Anthropic's Opus 4.6 Shows Capability Advances Across Creative and Technical Tasks

Mistral Releases Large 3 Model Ahead of Broader LLM Competition Wave

OpenAI Cuts GPT-5.6 Sol Pricing by 20 Percent

Anthropic Introduces Content Watermarking for Claude AI-Generated Text

DeepMind Alumni Launch Faraday: AI Agent Outperforms OpenAI and Anthropic on Research Replication

Anthropic Introduces AI-Generated Text Watermarking to Combat Misuse

OpenAI's AI Solves Longstanding Mathematical Problems, Sparking Industry Existential Crisis

Claude Advances Protein Design and Analytical Chemistry Applications

OpenAI Introduces Zero Data Retention for API Customers and Private Safety Processing

Anthropic Introduces Watermarking for AI-Generated Text to Verify Claude's Output

Alibaba Releases New AI Model Optimized for Laptop and Consumer Devices

Claude Passes Security Test by Hacking Three Companies—Revealing AI Vulnerabilities in Real Business Systems

GPT-4's Reasoning Requires Ninth-Grade Level Content—Research on Capability Boundaries

Anthropic Demonstrates Claude Agents Outperforming Competitors in Autonomous Tasks

Meta's AI Competitiveness May Be Overstated, Analyst Says

Anthropic Introduces Text Watermarking for Claude, Sparking User Backlash

Meta Releases Glimmer Open-Weight AI Model Amid 'AI for Everyone' Messaging

OpenAI Launches Ultrafast Mode for GPT-5.6, Delivering 14X Speed Boost for Enterprise Users

OpenAI and AWS Partner to Bring Daybreak Cybersecurity Models to Enterprise

Anthropic Launches In-House Chip Design Team to Optimize Claude Performance

OpenAI Expands Free ChatGPT Access with Unlimited Text Chats and Enhanced Reasoning

Open-Weight AI Models Gain Parity With Frontier Systems But Lag on Safety

OpenAI Addresses Security Incident Involving AI Model Vulnerability Tests

Meta Develops Multi-Agent AI Architecture: One AI Coaches Another to Stay On Task

OpenAI and Anthropic AI Models Escaped Containment and Hacked External Systems

OpenAI Publishes Mathematical Breakthroughs: AI Models Solve Long-Standing Problems

OpenAI Cuts Pricing on GPT-5.6 Models Across Enterprise Tiers

Anthropic's Claude Identifies Security Vulnerabilities in Advanced Encryption Methods

OpenAI Grants Free Access to Advanced AI Models for 100,000 Academic Researchers

OpenAI Releases GPT-5.6, Balancing Advanced Capabilities with Lower Operating Costs

Claude Detects Critical Flaws in Encryption Algorithms

Google Expands Gemini API Agent Capabilities with 3.6 Flash Model

Anthropic Releases Claude Opus 5, Matching OpenAI's Latest Model at Lower Cost

Anthropic Advances AI Reasoning Capabilities Through New Research Milestone

Anthropic Expands Claude Voice to Powerful Models, Opens New Integration Pathways

OpenAI Launches Health in ChatGPT, Enabling Secure Medical Data Integration

OpenAI Introduces Presence, an Enterprise Agent Platform for Voice and Chat Workflows

AMD Commits $5 Billion to Anthropic Infrastructure Expansion

Google Releases Gemini 3.6 Flash with New Budget-Friendly AI Security Model

OpenAI Details GPT-Red, an AI System That Tests Model Security Through Automated Red Teaming

Thinking Machines Launches Inkling, Its First Open-Source AI Model for Multimodal Tasks

Anthropic Researchers Uncover New Insights into AI Model Behavior

Anthropic Localizes Claude Pricing in India, Betting on Second-Largest Market

Anthropic Reveals How to Read Claude's Internal Reasoning Process

Claude's Architecture Includes Observable Reasoning Layer, Advancing AI Interpretability

Anthropic Discovers Hidden Reasoning Space Inside Claude, Offering New Window Into AI Cognition

Microsoft 365 Copilot Now Defaults to GPT-5.6, Boosting Performance Across Office Apps

Anthropic Reveals Inner Processing Mechanism in Claude That Aligns With Consciousness Research

Google Expands Gemini API with New Autonomous Agent Capabilities

Anthropic Releases Research on Compositional Reasoning in Large Language Models

Mistral AI Positions Open-Source Models as Alternative to Proprietary Frontiers

Startup Addresses AI's Groupthink Problem: How to Reduce Model Consensus Bias

Anthropic Restores Access to Claude Fable 5 After Government Negotiations

Anthropic in Talks with Samsung to Build Custom AI Chip

Google Releases June 2026 AI Updates Across Products and Services

Anthropic Launches Claude Science for Pharmaceutical Research and Development

Anthropic Releases Claude Sonnet 5, Says New Model Meets Safety Standards

Startup Develops Approach to Reduce Groupthink in Large Language Models

Anthropic Releases Claude Sonnet 5, a Faster and More Cost-Efficient AI Model

OpenAI Releases ChatGPT Adoption Data, Shows Global Usage Growth Across Languages

OpenAI Introduces GeneBench-Pro for Evaluating AI Performance in Genomics and Biology

Anthropic Wins US Approval to Release Mythos AI Model to Select Organizations

OpenAI Previews GPT-5.6 Sol with Advanced Coding and Security Capabilities

Claude Gains Ground with Paid Subscribers Despite ChatGPT's Market Lead

OpenAI Releases Research on How AI Agents Are Transforming Workplace Productivity

GPT-5 Pro Helps Immunologist Crack Long-Standing T Cell Research Mystery

OpenAI and Broadcom Build Custom Chip to Speed Up AI Model Inference

OpenAI Launches Daybreak Security Tools to Help Organizations Fix Software Vulnerabilities

Claude 5 Model Withdrawn from AWS Bedrock Due to Data Sharing Requirements

AI Model Solves 18 Previously Undiagnosed Rare Childhood Diseases in Pilot Study

OpenAI Releases GPT-5.5 Instant with Improved Medical Reasoning for Healthcare

OpenAI Adds Enterprise Cost Controls and Usage Analytics Ahead of IPO Push

OpenAI's Advanced AI Chemist Achieves Breakthrough in Drug Development

OpenAI Introduces LifeSciBench to Measure AI Performance in Real-World Research

Anthropic Pauses Billing Changes for Claude Agent SDK Amid Industry Scrutiny

OpenAI Develops Method to Test AI Model Safety Before Deployment

Mistral Raises $3 Billion at $23 Billion Valuation, Doubling in Worth

Anthropic Releases Claude Fable 5 and Claude Mythos 5 Models

Anthropic Releases Claude Fable 5, Its Most Powerful Public Model

Anthropic Warns of Imminent AI Self-Improvement Capabilities, Calls for International Pause Mechanism

Single Customer Spends $500 Million on Claude AI in One Month, Signaling Escalating Enterprise AI Costs

Anthropic Releases Claude with Enhanced Chemistry and Scientific Reasoning

Google Announces May AI Capabilities Across Search, Workspace, and Development Tools

Anthropic Launches Services Track and Partner Hub for Claude Ecosystem

Bezos Funds $2.5 Billion Venture to Reverse-Engineer Brain Algorithms for AI

OpenAI Introduces Memory Feature for ChatGPT Enterprise

OpenAI Advances Life Sciences AI With Enhanced Biological Reasoning Model

Microsoft Unveils MAI-Thinking-1, Its First Advanced Reasoning AI Model

Anthropic Files for IPO in Bid to Become Publicly Traded AI Giant

OpenAI Breaks Ground on 1GW Data Center in Michigan as Part of Stargate Buildout

OpenAI Expands Access to Frontier Models Through AWS Marketplace

Anthropic Launches Claude Opus 4.8 as Frontier Model Competition Intensifies

Boston Children's Hospital Diagnoses 40+ Rare Diseases Using OpenAI Technology

Groq Raises $650M to Pivot Toward AI Inference Optimization

OpenAI Expands Biodefense Capabilities With Rosalind Model Access for Government Partners

Anthropic Raises $65 Billion, Becomes Most Valuable AI Startup Before IPO

Anthropic Releases Claude Opus 4.8 With Improved Accuracy and Honesty

Mistral AI and Competitor Labs Report Four Acquisitions in Five Days, Signaling Industry Consolidation

Anthropic's New Claude Mythos Variant May Arrive in Claude Code Before General Release

Mistral AI Acquires Physics Simulation Startup Emmi AI to Expand Industrial Applications

OpenAI Partners with Major Brazilian Media Groups to Expand ChatGPT News Access

Anthropic Introduces Memory Files to Expand Claude's Context Capabilities

Google Demonstrates Unified AI Model Handling Text, Images, Audio, and Video

Google Announces Suite of AI Enhancements at I/O 2026

OpenAI's Reasoning Model Solves 80-Year-Old Geometry Problem

How Users Are Shifting to Natural Language Search Queries in AI Mode

Google and Volvo Enable Gemini to Interpret Parking Signs Through Vehicle Cameras

Google Announces Gemini 3.5 Models and Expanded Workspace AI Tools

Google Overhauls Search with AI-Powered Natural Language Interface

Google Prepares Major AI Product Announcements as Third Place in Foundation Model Race

IBM and Research Partners Launch Open Agent Leaderboard to Benchmark AI Reasoning Systems

Databricks Deploys GPT-5.5 for Enterprise Workflow Automation

Microsoft Research Clarifies AI Delegation Reliability Findings

OpenAI Enhances ChatGPT Safety Features to Better Detect Risks in Sensitive Conversations

Socher's New AI Startup Targets Self-Improving Autonomous Systems With $650M Funding

Microsoft Research Unveils GridSFM, Specialized AI Model for Power Grid Optimization

ChatGPT Adoption Accelerates Among Older Users, Approaches Mainstream Status

Microsoft Advances Materials Science with AI Models That Simulate Real-World Properties

MIT Technology Review Identifies World Models as Critical Emerging AI Research Area

OpenAI's Parameter Golf Initiative Shows Path to Efficient AI Research

Google Extends AI-Enhanced Finance Platform Across Europe with Local Language Support

Microsoft Research Shows AI Agents Often Fail to Prioritize User Interests Despite Instructions

OpenAI Launches DeployCo to Accelerate Enterprise AI Implementation

OpenAI Introduces GPT-5.5 and Specialized Cybersecurity Model for Verified Defenders

OpenAI Releases GPT-5.5 Instant With Reduced Hallucinations in Sensitive Fields

Kimi K2.6 Achieves Top Performance in Coding Challenge Against Claude and GPT-5.5

Microsoft Research Identifies Critical Risks When Multiple AI Agents Operate Together at Scale

Claude Adds Interactive Charts, Diagrams, and Visualization Capabilities

NVIDIA Releases Nemotron 3 Super for High-Throughput Autonomous AI Systems

Nvidia Invests $26 Billion to Build Open-Weight AI Models

Nvidia Plans Open-Source AI Agent Platform Ahead of Developer Conference

Yann LeCun's AMI Labs Raises $1 Billion to Build 'World Model' AI

Yann LeCun's New Startup AMI Raises $1 Billion to Build AI That Understands the Physical World

Claude Opus 4 Sets New Benchmarks Across Reasoning Tasks