X Platform Regains Stability Under Elon Musk's Ownership
John Gruber wrote in Daring Fireball that X has regained its footing under Elon Musk's ownership, with discussions on AI and other interesting topics on the platform being significantly higher in quality and activity compared to rival platforms like Threads, Bluesky, and Mastodon.
Apple's new CEO John Ternus posted on X shortly after taking office last month, with content including "hello", previews before the iPhone event, new product promotional videos, Emmy award celebration posts, Fifth Avenue launch day records, and a tribute to Steve Jobs on the anniversary of his passing. These posts appeared exclusively on X, and marketing head Greg Joswiak's account also only updates on X. Tim Cook even posted about Ternus interacting with Pikachu.
Gruber noted that hundreds of interesting accounts he follows continue to post only on X and not on other platforms. For example, the Merriam-Webster dictionary account has published 45 original and retweeted posts on X since October, while Bluesky has only 10. New applications showcasing and discussing topics are also concentrated on X.
The platform culture presents a friendly, inclusive, and pleasant atmosphere, with users receiving "welcome back" messages rather than blame upon returning. Discussions about AI tools and use cases can occur directly, unlike Bluesky and Mastodon, which are often interrupted by anti-AI sentiments. X's algorithm has been depersonalized, no longer forcing the promotion of Musk's personal posts, and the timeline is primarily focused on content users follow, with some users posting complete articles and blogs.
If someone returns after a year away, they might think Musk has grown tired of and sold the platform, but in reality, after reducing micro-management, the strong personal touch from the early takeover has faded, and the density of actions and information is significantly higher than on other networks.
Advertisers and content creators are returning to X, as institutional users and decision-makers are concentrated here. Event-driven executive posts and algorithm adjustments reinforce the network effect, directing funds toward platform advertising and subscriptions, benefiting tech companies and media that rely on real-time discussions, while decentralized alternative platforms face pressure.
X maintains hundreds of millions of monthly active users. Although advertising revenue has not fully recovered to pre-takeover levels, most top advertisers have returned, and the transparency from the open-sourced algorithm has further attracted professional discussants.
Source: Public Information
ABAB AI Insight
After Elon Musk acquired Twitter for $44 billion in 2022, he implemented massive layoffs down to about 1,800 employees, rebranded it as X, and due to relaxed content moderation, advertising revenue plummeted. After hitting a revenue low in 2023, the platform was integrated into the rocket and AI business systems through xAI and SpaceX acquisitions, with its valuation temporarily dropping to the $30-40 billion range. Investors, including Larry Ellison and Andreessen Horowitz, gained SpaceX shares through equity conversion.
In terms of capital strategy, Musk injected X's equity into SpaceX, using the latter's IPO expectations to dilute debt and attract advertising back. In 2026, he plans to offer returning advertisers up to $200,000 in additional value incentives while open-sourcing the recommendation algorithm based on the Grok transformer architecture. Resources are shifting from content moderation to AI data supply and real-time training, aiming to transform the social network into a user entry point and data pipeline for xAI and SpaceX.
This is similar to Facebook's recovery growth through algorithm adjustments and advertiser appeasement after the 2018 Cambridge Analytica incident, and akin to YouTube consolidating its dominance through scale and recommendation engines after early content controversies. Currently, X is in a control phase after expansion, locking in institutional and tech users through network effects rather than mere scale expansion.
Essentially, this represents capital concentration and industrial chain restructuring: social data and AI model training are merged under SpaceX, with pricing power shifting from traditional advertisers to platform data value. The irreplaceable network lock-in arises from real-time discussions and high executive content density, with mechanisms of scale inertia and algorithm depersonalization reducing friction, making it difficult for alternative platforms to replicate action density.
ABAB News · Cognitive Law
- Network effects outweigh aversion, locking in actions by scale.
- Algorithm depersonalization equals platform rebirth.
- Data pipelines are worth more than content itself.