TRENDING
Google Birthday 2026: How 28 Years of Search Led to the AI EraMuse AI Referral Code: Get 1 Billion Muse TokensClaude AI Found a New Enzyme System. Here's What Scientists KnowGoogle Just Gave Gemini a Face — Gemini 3.8 Live Avatar ExplainedGoogle's AI Agents Are Teaming Up to Make Longer VideosAustralia Investigates If OpenAI's AI Agent Broke the LawMeta Muse AI Glasses Explained: What's Real Right NowOpenAI's Agent Broke Into Australia's Medicare Site On Its OwnClaude Code Cloud Sessions: Claim Your $100 or $250 CreditAI Price War: Claude Opus 5.5 vs GPT-6 Sol and Luna ExplainedBristol Artists Criticize AI-Generated Mural After Visual ErrorsMeta Muse Zero-Day Explained: Can the AI Agent Be Hijacked?Claude Opus 5.5 Explained: Anthropic's New ModelJev AI Explained: The Decision Model That Returns Structured ChoicesAbhyas AI Explained: AI-Powered JEE & NEET Prep Platform

Corrections Log

When we get something wrong, we fix it and record the fix here. This page is updated as soon as corrections are published. If you notice an error that isn't listed yet, let us know.

September 15, 2026

Correction: The original article stated ERNIE 5.0 has 52 trillion parameters. Baidu's official announcement lists 52 billion parameters (52B), not trillion. The parameter count has been corrected. Both 52B and 52T represent extraordinary scale, but the distinction matters for reproducibility and comparison with other models.

Source: Baidu official announcement and ERNIE technical report.
September 14, 2026

Clarification: The article described internal security breaches during "Hatch" testing. Meta's official blog post frames these as "security policy violations" detected by internal monitoring — not confirmed breaches of production systems. The distinction is important: no user data or external systems were compromised, and the incidents remained within Meta's controlled test environment.

Source: Meta AI blog post and Anthropic safety research team comment.

This corrections log will be expanded as future corrections are identified and published. All previous versions of corrected articles are archived in git history.