Alibaba has introduced stronger safety measures for its latest artificial intelligence (AI) model as Chinese technology companies face growing pressure to explain how they test and protect advanced AI systems. The company says its new Qwen4-Max model includes additional safeguards against cyberattacks and other harmful uses. The move reflects a wider shift in China’s AI industry, where developers are paying greater attention to safety testing, transparency, and responsible technology development.
Alibaba Introduces New Safety Measures for Qwen4-Max
Alibaba, one of China’s largest technology companies, has announced new security features for its flagship AI model, Qwen4-Max. The model is designed to support complex professional and scientific tasks while reducing the risk of misuse.
According to the company, its training process allows the model to retain cybersecurity knowledge that can help developers identify and fix software weaknesses. However, it excludes training material intended to improve offensive hacking capabilities.
This approach aims to support defensive cybersecurity activities without making the technology more useful for cybercrime.
Alibaba has also strengthened protections against risks involving chemical, biological, radiological, and nuclear threats. The company said its internal evaluations found no evidence that the model could significantly help users carry out dangerous activities. These findings have not been independently verified by Reuters.
Improved Performance and Lower Operating Costs
Along with safety improvements, Alibaba has highlighted the technical performance of Qwen4-Max.
The company reported that the model achieved better results in secure code generation compared with its previous version.
Key developments include:
- Secure code generation: Qwen4-Max scored 14.7% on an internal benchmark, compared with 8% for its predecessor.
- Lower operating costs: Alibaba reported an approximately 60% reduction in average task costs during internal testing.
- Advanced capabilities: The model contains 2.4 trillion parameters and supports complex professional and scientific applications.
- Future development: Alibaba plans to introduce smaller open-weight models from the Qwen4 family, allowing developers to download and modify them.
- These improvements show how AI companies are working to balance model performance, affordability, and safety.
Chinese AI Developers Increase Safety Disclosures
Alibaba’s announcement comes as Chinese AI developers face increasing questions about transparency and safety testing.
An October 2026 report by research firm SemiAnalysis reviewed 857 AI model releases from nine major Chinese companies between 2021 and September 15, 2026. It found that only 31 releases, or 3.6%, had publicly available safety evaluation results linked to specific models.
The report also found that just nine models had published safety results before or at launch.
Following these findings, companies including Tencent and Moonshot AI have also announced additional restrictions on cybersecurity-related capabilities in their AI systems.
The developments indicate growing industry attention toward sharing clearer information about how AI models are evaluated and protected.
Why AI Safety and Transparency Matter
As artificial intelligence becomes part of business operations, healthcare, finance, education, and cybersecurity, protecting users from potential misuse is becoming increasingly important.
Safety testing helps developers identify weaknesses before AI systems reach wider audiences. Public disclosures also allow businesses, researchers, and users to better understand the capabilities and limitations of these technologies.
However, stronger safeguards must continue to develop alongside increasingly powerful AI models.
Frequently Asked Questions
1. What is Alibaba’s Qwen4-Max?
Qwen4-Max is Alibaba’s flagship AI model, designed for professional, scientific, and other complex tasks.
2. What new AI safeguards has Alibaba introduced?
The company has strengthened cybersecurity protections and added safeguards against dangerous chemical, biological, radiological, and nuclear applications.
3. Why are Chinese AI developers increasing safety disclosures?
Growing concerns about AI misuse and limited public information about model testing have increased pressure on developers to provide greater transparency.
4. What is the importance of AI safety testing?
AI safety testing helps identify potential risks, reduce harmful outputs, and improve the responsible development of artificial intelligence.
Building a Safer and More Transparent Future for AI
Alibaba’s latest AI safety measures highlight an important development in China’s artificial intelligence industry. As models become more powerful and widely accessible, responsible development, security testing, and transparent reporting will play an increasingly important role. The next stage of AI growth will depend not only on building advanced systems but also on ensuring they can be used safely and responsibly.
Read our latest interview with Dr. Huang Wei Ling

