Yazılar

Alibaba’s Amap Enters Local Business Rankings, Challenging Meituan

/in /tarafından

Alibaba’s mapping app Amap is expanding beyond navigation by launching AI-driven rankings for restaurants, hotels, and tourist destinations, directly competing with Meituan’s Dianping platform in the lucrative local-lifestyle services market.

Key Details

  • Street Stars Launch: Amap’s new feature, called Street Stars, ranks local businesses for its 170 million daily active users.

  • Financial Incentives: Alibaba is offering 1 billion yuan ($140 million) in subsidies for coupons on ride-hailing and in-store services.

  • Coverage: The service initially spans 300 cities and 1.6 million local business listings.

Competitive Landscape

  • Meituan’s Response: Meituan countered with 25 million coupons tied to top-rated restaurants, signaling an intensifying rivalry.

  • Instant Retail Battle: Both firms are locked in a discount-heavy price war in China’s booming one-hour delivery and lifestyle services segment.

  • CEO Vision: Alibaba CEO Eddie Wu described Amap’s AI transformation as a step toward making it a “gateway for future lifestyle services.”

Regulatory Angle

  • Scrutiny Rising: Chinese regulators have repeatedly warned against race-to-the-bottom pricing strategies and have summoned top firms for discussions.

  • Economic Context: The competition comes as consumer spending remains weak due to housing market troubles and job insecurity, pushing platforms to rely on aggressive subsidies.

Implications

Amap’s pivot signals Alibaba’s intent to build a comprehensive consumption ecosystem, but regulators’ intervention could determine how far the subsidy-driven price war continues in China’s local services and lifestyle economy.

TikTok building U.S.-only app with separate algorithm and data systems

/in /tarafından

TikTok is developing a standalone U.S. version of its platform, complete with a distinct algorithm and data system, to comply with U.S. legislation that mandates the divestment of its American operations. The project, internally known as “M2,” aims to meet a September deadline and could clear the path for a potential sale of TikTok’s U.S. business, Reuters reports, citing employees with direct knowledge.

The move involves duplicating TikTok’s codebase — including AI models, algorithms, features, and U.S. user data — from its global app to an independent U.S.-specific version. It is TikTok’s most ambitious technical separation effort to date and would represent the deepest structural divide between ByteDance’s U.S. and international operations. The U.S.-only version would function much like Douyin, TikTok’s China-specific app, and would not be visible to users outside the U.S.

The initiative responds to the 2024 law requiring ByteDance to divest TikTok or face a ban, amid long-standing U.S. concerns about data privacy and national security. While content from the current app is expected to carry over, the new recommendation engine will be trained solely on U.S. user data. This is expected to shift content visibility toward American creators and possibly limit international reach for non-U.S. influencers.

Sources revealed that since January, TikTok has been removing non-U.S. user data from Oracle’s American data centers to comply with separation demands. Meanwhile, ByteDance has worked on splitting its algorithm’s codebase — a move it previously denied.

If the technical split is completed, U.S. operations would be managed independently of TikTok’s global team, although ByteDance engineers might remain involved on a limited basis. This has raised internal questions about whether the U.S. algorithm will retain its effectiveness without access to ByteDance’s global engineering expertise.

A potential sale would involve a joint venture including American investors such as Susquehanna International Group, General Atlantic, KKR, and possibly Oracle, along with new players like Blackstone and Andreessen Horowitz. ByteDance would retain a minority stake. However, Beijing’s approval remains uncertain due to China’s export restrictions on recommendation algorithms, a key concern in the stalled 2020 negotiations.

The separation effort is unfolding against a broader backdrop of U.S.-China trade tensions. While former President Donald Trump said last week he would resume discussions with China over the sale, he admitted uncertainty over Beijing’s cooperation, adding, “I think the deal is good for China and it’s good for us.”