Hands on with X’s new AI-powered custom feeds
X is rolling out Grok-powered custom feeds to replace its Communities feature, integrating AI curation directly into timeline personalization while adding new advertising inventory. The shift signals X's strategy to embed its LLM deeper into core user experience and monetization.
Modelwire context
Analyst takeThe replacement of Communities, a social-organizing feature, with an AI-curated feed is a quiet product burial that deserves more scrutiny than the Grok integration itself. X is trading user-controlled community structure for algorithmic curation it can monetize, which is a different value proposition than what Communities promised.
The pattern here mirrors what we covered around Gemini's Google Photos integration (Ars Technica, April 16): major platforms are routing their LLMs through existing user-data surfaces to deepen stickiness and justify the model's presence in daily product flows. The difference is that Google's move was additive, while X's is substitutive, retiring a feature to make room for the AI layer. That distinction matters for how users and advertisers respond. The broader context from the tokenmaxxing coverage (TechCrunch, April 17) is also relevant: as AI infrastructure costs compound, platforms need LLMs to pull direct revenue weight, not just serve as features.
Watch whether advertiser CPMs on Grok-curated feeds outperform X's standard timeline inventory within the next two earnings cycles. If they do, expect Meta and YouTube to accelerate similar LLM-to-feed integrations; if they underperform, the monetization rationale for retiring Communities collapses.
Coverage we drew on
- Gemini can now create personalized AI images by digging around in Google Photos · Ars Technica — AI
This analysis is generated by Modelwire’s editorial layer from our archive and the summary above. It is not a substitute for the original reporting. How we write it.
MentionsX · Grok · TechCrunch
Modelwire Editorial
This synthesis and analysis was prepared by the Modelwire editorial team. We use advanced language models to read, ground, and connect the day’s most significant AI developments, providing original strategic context that helps practitioners and leaders stay ahead of the frontier.
Modelwire summarizes, we don’t republish. The full content lives on techcrunch.com. If you’re a publisher and want a different summarization policy for your work, see our takedown page.