Alibaba’s Qwen App is testing paid features, according to a report carried by Pandaily, putting the company’s consumer AI product into a monetization experiment as Chinese AI platforms look for ways to convert heavy usage into revenue. The move is being discussed alongside ByteDance’s Doubao, whose use of an office-focused product strategy has become a reference point in the market.

A separate 36Kr report frames the question more directly: after Qwen explored paid functionality, can Doubao successfully apply paid feature monetization itself? The available source material does not specify which Qwen features are being tested, whether users are already paying, or whether Doubao has formally launched a comparable paid plan. Those gaps make the development more significant as a strategic signal than as proof of a completed business model.

What the reports establish

The clearest reported event is that Alibaba’s Qwen App is testing paid features. Pandaily’s headline describes the test as part of an effort to follow what it calls Doubao’s “office playbook,” while 36Kr presents the development as part of a broader debate over Doubao’s ability to monetize features.

The reports do not provide enough detail to establish the test’s scope. There is no confirmed information in the supplied evidence about pricing, eligibility, payment conversion, feature limits, geographic availability, or whether the experiment is available to all Qwen users. It is therefore too early to describe Qwen’s move as a full subscription launch or to conclude that Alibaba has found a repeatable revenue model.

The same caution applies to Doubao. The source titles connect Doubao with office-oriented product positioning and paid feature monetization, but the evidence does not confirm a specific Doubao pricing announcement. Any claim that Doubao has already achieved meaningful paid adoption would go beyond the available reporting.

Why Qwen and Doubao are being compared

The comparison matters because it places two major Chinese AI apps in the same product question: which capabilities can be valuable enough for users or organizations to pay for? Free access can help an AI assistant build a large user base, but it also creates pressure to cover model inference, storage, support, and product development costs.

An office-oriented strategy offers one possible route to payment. Instead of treating an AI app only as a general-purpose chatbot, a provider can attach value to recurring work such as drafting, document handling, research, collaboration, or other productivity tasks. The source evidence does not identify which office functions Doubao emphasizes or which Qwen capabilities are behind the paid-feature test, so the precise product comparison remains unresolved.

Still, the framing from 36Kr suggests a competitive sequence: Qwen is experimenting with paid functionality, while Doubao’s office strategy is being evaluated for whether it can support similar monetization. That makes the story less about a single feature release and more about whether Chinese AI companies can move from broad consumer distribution to dependable commercial usage.

Evidence, claims, and important limits

Both available items are media reports distributed through Google News links, and the supplied extracts contain only headlines and summaries. There are no official Alibaba or ByteDance statements, product documentation, pricing pages, user data, or independently verified performance figures in the evidence provided here.

Accordingly, the paid-feature test should be treated as a reported product experiment rather than a confirmed revenue milestone. “Office playbook” is a media characterization, not a documented description of a formal Doubao business plan in the supplied material. Likewise, the 36Kr question about Doubao’s prospects is market analysis, not evidence that the company has announced or validated a paid offering.

This distinction is important for builders and enterprise buyers. A feature appearing behind a payment prompt does not by itself demonstrate willingness to pay. Teams evaluating an AI product need to know whether the paid capability improves a measurable workflow, whether usage limits are predictable, and whether the provider can maintain service quality as demand grows. None of those questions is answered by the reports available here.

Implications for AI products and buyers

For AI product teams, Qwen’s reported test is a reminder that monetization depends on packaging, not simply model access. A provider must decide which functions remain broadly available, which capabilities justify payment, and whether the value is best captured through individual subscriptions, team plans, usage-based charges, or enterprise contracts. The comparison with Doubao suggests that office workflows may be an important proving ground, but the available evidence does not show which pricing structure either company prefers.

For enterprises, the immediate lesson is to separate product experimentation from procurement readiness. A paid feature may indicate that a vendor is testing commercial demand, but buyers still need clarity on data handling, administrative controls, integration options, reliability, support, and contract terms. Those issues become especially important when an AI assistant is tied to workplace documents or recurring business processes.

For founders and researchers, the development points to a crowded competitive layer beyond model quality. Qwen and Doubao may be competing not only on answer quality or usage scale, but also on how effectively they turn an assistant into a repeatable work tool. If paid features are limited to novelty functions, conversion may remain weak. If they remove costly bottlenecks in a real workflow, they could create a stronger basis for recurring revenue. The sources do not yet reveal which path is emerging.

What to watch next

The most useful follow-up signals are concrete product and business disclosures. First, Alibaba’s Qwen App would need to clarify which paid features are being tested, who can access them, and whether the test has a stated price or billing model. Any expansion from a limited experiment to a public plan would provide stronger evidence of commercial intent.

Second, the market will need a confirmed Doubao announcement, if one occurs, describing its office products, paid capabilities, and target customers. A formal launch would make it possible to compare the two services on workflow coverage and pricing rather than on media framing alone.

Third, independent indicators will matter more than feature announcements: sustained usage of paid tools, renewal behavior, enterprise deployments, and evidence that customers use the products for recurring work. Neither Pandaily nor 36Kr, in the supplied material, provides those measurements.

Creati.ai perspective

The reported Qwen test is important because it puts monetization on the product roadmap, but it is not yet proof that paid AI features work at scale. The strongest conclusion supported by the evidence is that Alibaba is exploring payment while Doubao’s office positioning is being judged against the same commercial challenge.

For the AI market, the next meaningful milestone will not be another paid toggle. It will be evidence that a feature solves a frequent, expensive workflow well enough for users or organizations to keep paying. Until Qwen or Doubao provides clearer product, pricing, and adoption data, the competition remains an open test of packaging and execution rather than a settled monetization model.