← Back to archive
Xbox's AI boss bans AI slop, guards human creativity

Image via The Atlantic

Wednesday, February 25, 2026

Xbox's AI boss bans AI slop, guards human creativity

We're watching an AI executive take the helm at Xbox with a promise to avoid "AI slop" and keep games human-crafted (bold move in 2026), while Google quietly launches ProducerAI for music creation in Labs. But here's where it gets messy: Chinese AI labs just got caught using fraudulent accounts to steal Claude's capabilities through distillation attacks, and schools are scrambling as AI tutors replace teachers with reports of real learning gaps and student distress (yikes). Meanwhile, Qwen drops an efficient 35B model with 1M context that actually works, proving innovation doesn't always need to be flashy. So the real question: if AI can now steal from AI, who's actually innovating anymore?

Top Stories

1
Xbox Is Run by an AI Exec Now

Microsoft's new AI-focused Xbox CEO promises to balance innovation with human creativity, rejecting AI-generated content while integrating AI strategically into gaming's future. The leadership change signals Microsoft's approach to gaming evolution in the age of AI.

microsoftgamingaileadership
2
AI Teacher Inside Alpha School

Wired

Alpha School's AI-driven learning model, backed by billionaire tech leaders, is under scrutiny after former students and parents report educational gaps, psychological harm, and inadequate instruction despite the school's marketing claims of academic success. The expansion of this surveillance-heavy, metrics-focused approach raises concerns about whether technology-enabled education can replace qualified teachers.

ai-in-educationedtechpersonalized-learningmicroschools
3
AI Innovation Is Starting to Look a Bit Recycled

Anthropic

Anthropic exposed coordinated distillation attacks by three Chinese AI labs that extracted Claude's capabilities through millions of fraudulent API calls, highlighting how competitors can bypass export controls and acquire frontier AI without necessary safety safeguards—a threat requiring industry-wide coordinated response.

anthropicexport-controlssecuritydistillation
4
Qwen3.5-35B-A3B

Hugging Face

Qwen3.5-35B-A3B advances multimodal foundation models with efficient sparse architecture, extended context windows up to 1M tokens, and native support for agentic reasoning across 201 languages, delivering competitive performance with reduced computational overhead.

llmmultimodalopen-sourceagents
5
ProducerAI Joins Google Labs

Google

Google Labs now hosts ProducerAI, a generative AI music creation tool powered by advanced DeepMind models, designed to democratize music production for artists at all skill levels while maintaining focus on creative control and human artistry.

generative-aimusicgooglecreative-tools

Keep Reading

Industry Voices

Enjoyed this issue?

Get daily AI intel delivered to your inbox. No fluff, just the stories that matter.