# AI News Digest - March 5, 2026

## 1. Pentagon Designates Anthropic as Supply Chain Risk
**Source:** The Verge
Defense Secretary Pete Hegseth has designated Anthropic as a "supply chain risk," leading defense contractors to pivot away from Claude AI. Companies are abandoning Claude preemptively, significantly impacting Anthropic's enterprise market.
**Relevance:** 9/10

## 2. OpenAI Developing GitHub Rival
**Source:** The Verge
Prompted by recent GitHub outages, OpenAI is developing its own code repository platform — putting it in direct competition with Microsoft, which holds a significant stake in OpenAI.
**Relevance:** 8/10

## 3. ChatGPT Gets GPT-5.3-Instant Update
**Source:** The Verge
OpenAI rolled out GPT-5.3-Instant with improved accuracy and context understanding, reducing unnecessary caveats and overbearing tone.
**Relevance:** 7/10

## 4. Google AI Ultra Gets Video Generation
**Source:** The Verge
Google AI Ultra subscribers can now turn notes into personalized, fully animated videos — Google's push into AI-powered creative content generation.
**Relevance:** 7/10

## 5. AI Wikipedia Translations Causing Hallucination Problems
**Source:** The Verge
The Open Knowledge Association's AI-translated Wikipedia articles contain hallucinations including fabricated sources. Wikipedia editors are now restricting OKA contributors.
**Relevance:** 6/10

---
**Theme:** Regulatory scrutiny (Anthropic), strategic competition heating up (OpenAI vs Microsoft), and ongoing AI reliability challenges.