Chinese tech giant Baidu took a major step on Wednesday to redefine its core services, launching a new AI video generator, MuseSteamer, alongside the most significant overhaul of its search engine in a decade. Announced from Beijing, the twin releases leverage Baidu’s powerful Ernie AI to transform its platforms.
The goal is to shift from simple information tools into interactive, task-oriented assistants. This strategic push aims to counter fierce competition from domestic rivals like ByteDance and global players such as Google. It also builds on Baidu’s recent, aggressive pivot to open-source AI.
This signals a clear intent to compete on both capability and cost in the escalating global AI race.
From Search Box to Smart Assistant
Baidu’s search platform has undergone its most sweeping change in years. The company has enlarged the mobile app’s search bar, turning it into a “smart box” powered by its Ernie AI models. This new interface supports long text inputs of thousands of words, a significant leap from previous limits.
The upgrade expands search beyond simple text queries. It now handles voice and image-based searches and integrates over 18,000 third-party AI agents. This allows the platform to perform complex tasks like planning trips, generating AI images, and even writing research reports.
This evolution reflects Baidu’s ambition to expand the “boundaries of search,” moving from a tool that retrieves information to one that helps users complete tasks directly within the app. The changes are being rolled out in phases over the coming weeks.
MuseSteamer Enters the AI Video Arena
Simultaneously, Baidu unveiled MuseSteamer, an image-to-video model aimed squarely at business users. The tool can generate video clips up to 10 seconds long and is available in three versions: Turbo, Pro, and Lite. Unlike consumer-focused rivals, Baidu has not yet released a public-facing app.
To promote its new tool, Baidu launched a creative contest on its Huixiang platform, encouraging users to generate “pinch-and-play” videos, a popular meme style in China. This move thrusts Baidu into a crowded field against global tech titans like Google, Microsoft, and Meta.
The companies have all recently launched or upgraded their own video generation tools, such as Google’s Veo 3 and Microsoft’s free, Sora-powered Bing Video Creator. The competition is fierce, with each player vying for dominance in a market projected to grow significantly.
A Calculated War on Price and Openness
These product launches are the latest salvos in Baidu’s aggressive new strategy. Just days ago, the company announced it was making its Ernie AI model family open source, a stunning reversal for a company that once championed proprietary systems.
This pivot is a direct challenge to the high-cost models of Western leaders like OpenAI. One analyst noted, “Baidu just threw a Molotov into the AI world… This isn’t a competition; it’s a declaration of war on pricing.” The war on pricing began earlier in the year, with Baidu making its Ernie Bot free and slashing API costs for its models by over 80%.
Baidu CEO Robin Li explained the strategy as a way to empower developers. “Why do we keep lowering the price of large language models? Because it’s a big obstacle for developers to make AI applications because of the high cost, they can’t afford it,” he stated at the company’s developer conference. This embrace of accessibility marks a profound shift from his previous skepticism about the open-source movement.
Geopolitical Chess and the Trust Deficit
Baidu’s open-source gambit is also a potent tool in the escalating U.S.-China tech war, potentially allowing it to circumvent American sanctions. However, this strategy is viewed with deep suspicion in Washington, where some see open-source AI from China as a national security risk.
U.S. officials have issued stark warnings. The House Select Committee on the CCP recently called rival Chinese AI firm DeepSeek a “weapon in the Chinese Communist Party’s arsenal, designed to spy on Americans, steal our technology, and subvert U.S. law.”. This highlights concerns about data privacy and the potential for misuse of the technology.
Experts caution that transparency remains a major issue. As Professor Sean Ren of USC warns, “Just because a model’s weights are public doesn’t mean we know what data it was trained on, whether consent was given, or if those data contributors were credited or compensated.”. This trust deficit could hinder Baidu’s ambitions, as it competes not just on features and price, but also on the promise of security and responsible development.