Latest
Apple Intelligence in China: Alibaba's Qwen and iOS 27.3
Apple Intelligence reaches China in December with iOS 27.3. Which features ship, what Alibaba's Qwen does in Siri, which iPhones qualify, and what's missing.
Read more →
AI Training Opt-Out in robots.txt: The New Shared Token
OpenAI, Google, Microsoft and Anthropic will honor one AI training opt-out in robots.txt: __usage_ai-training. Syntax, dates and what it misses.
Read more →
GPT-Rosalind-6: OpenAI's Biology Model, Access and Pricing
GPT-Rosalind-6 runs on GPT-6 Astra weights with fewer biology refusals. See benchmarks, $15/$75 pricing, the trusted-access rules and who can apply.
Read more →
Claude Code Extension Backdoor in Cursor: Check and Fix
Tampered Claude Code extension builds 2.1.312 and 2.1.313 in Cursor's marketplace planted an SSH backdoor on Oct 29. Who is affected and how to clean up.
Read more →
Alphabet Q3 2026 Earnings: Cloud Doubles, Cash Flow Negative
Alphabet Q3 2026 earnings: Google Cloud revenue rose 103% to $30.7B, but $51.4B of capex pushed free cash flow to minus $9.1B. The numbers and what they mean.
Read more →
Mistral Large 4 Open Weights: Specs, Benchmarks, License
Mistral released the weights of Large 4, a 1T-parameter MoE with 52B active. Specs, benchmarks vs Kimi K3 and GLM-5.3, license terms and how to self-host it.
Read more →
SK hynix Q3 2026 Earnings: 104.6T Won Revenue, 80% Margin
SK hynix Q3 2026 earnings: first quarter above 100 trillion won in revenue, an 80% operating margin on AI memory demand, and a 100 trillion won payout plan.
Read more →
Gemini 4 Carbon: Benchmarks, Pricing and Who Gets Access
Gemini 4 Carbon is Google's new top model for cybersecurity and long agentic tasks. Benchmarks vs Argon, GPT-6 and Claude, $2/$10 pricing and limited access.
Read more →
Liquid AI LFM3: 2B Open-Weight Models for On-Device AI
Liquid AI's LFM3 family brings text, vision, audio and omni models of about 2B parameters with open weights. Benchmarks, specs, memory use and license terms.
Read more →
Genie 4: Google DeepMind's World Model with Native Audio
Genie 4 is Google DeepMind's real-time world model that generates sound and dialogue with video. Who can use it, how it differs from Genie 3, and its limits.
Read more →
Loading more posts…