Have you heard these exciting AI news? - March 06, 2026 AI Updates Weekly

Quick Overview

The AI Updates for March 06, 2026, highlight significant advancements across model performance (Qwen3.5-9B, LFM2-24B-A2B), new tooling (Anthropic Skill Creator, Claude Code features like Memory and Voice Mode), major tech industry shifts (SaaS collapse prediction, AT&T's multi-agent savings), and competitive landscape changes (Anthropic vs. OpenAI revenue projections).

Key Points: Anthropic is predicted to surpass OpenAI in annualized revenue by mid-2026, with projections showing their growth trajectory crossing OpenAI's. New tools were announced, including the Anthropic Skill Creator tool for evaluating and optimizing skills, and Claude Code now supports Memory.md and a beta Voice Mode, which uses speech transcription without counting against token limits. Qwen3.5-9B was released by a Chinese team, offering strong performance on consumer GPUs with various quantizations (e.g., 4-bit requiring 5-6 GB VRAM) and multimodal capabilities. OpenAI's GPT-5.3 Instant is priced competitively at $1.75/$14 per 1M tokens for standard/long context, but deep reasoning is significantly more expensive at $38/$180. The presentation emphasizes a shift from specialized local apps/SaaS to generic AI agents ('Digital Employee') powered by skills, memory, and remote control. Harness Engineering is highlighted as a key trend, suggesting that the infrastructure wrapped around AI models (the 'harness') will matter more than the model itself, with 2026 predicted as the 'year of harnesses'. New local models like Gemini Nano Banana 2 offer significant speed and cost improvements over previous versions, and NullClaw demonstrates high efficiency (1MB vs over 1GB memory) due to being built in Zig.

Context: This weekly AI Update summary for March 06, 2026, covers the latest developments across foundation models, tooling, competitive dynamics, and emerging industry trends. Key announcements include new model releases from Qwen and updates to Anthropic's Claude Code features, alongside significant commentary regarding the competitive landscape between major AI labs like OpenAI and Anthropic, and a predicted shift in software architecture towards agentic systems.