Chinese technology company Alibaba has released a new version of its AI model, Kwen. The company claims that this version is more powerful than ChatGPT and the widely discussed Chinese AI DeepMind, which could potentially surprise the world.
Experts say this reflects the frantic rush to develop powerful new AI systems both inside and outside of China.
In an announcement, Alibaba’s Cloud division stated that Kwen 2.5 Max is superior to OpenAI’s GPT-4, DeepMind-V3, and Meta’s Llama-3.1-405B in almost every area. Meanwhile, analysts believe that Chinese AI models, like DeepMind, which are built at a lower cost but have high capabilities, could bring significant changes to the AI technology world. Furthermore, the cost-effective approach of Chinese companies in building AI models raises questions about the justification for massive investments in top American AI companies.
On January 10, the AI assistant powered by DeepMind-V3 was launched in the market. Then, on January 20, the startup released its AR1 AI model, which shocked Silicon Valley and caused a slump in technology sector stocks.
Chinese DeepMind’s AI model can provide similarly accurate answers while using 30% less power than OpenAI’s. This has raised questions among investors about the massive spending plans of leading AI companies in the United States.
With the success of DeepMind, intense competition has started among Chinese rivals to advance their AI models.
Just two days after the release of DeepMind’s AR1 model, TikTok’s parent company ByteDance released an update for their AI model, claiming that it defeated OpenAI’s O1 model in the AIME test.