In the face of rising competition in the artificial intelligence (AI) market, Chinese tech giant, Baidu BIDU, is preparing to roll out its next-generation AI model later this year.
What Happened: Baidu is set to release the next version of its AI model, Ernie 5.0, in the second half of 2025, reported CNBC on Thursday. This development comes amid the disruption caused by new entrants like DeepSeek in the AI industry.
The forthcoming model termed a “foundation model,” is expected to show significant improvements in multimodal capabilities, enabling the processing and transformation of various data types, including text, video, images, and audio. Chinese AI chatbots from startups and big-tech firms such as Alibaba and ByteDance have outpaced Baidu’s Ernie despite it being among the first ones to be launched.
The current Ernie model, Generation 4, was unveiled in October 2023, followed by an upgraded “turbo” version, Ernie 4.0, in August 2024. Meanwhile, in a WeChat post today, Baidu also stated that the existing chatbot will be available at no cost for individual users starting in April, reported Bloomberg. This will accompanied by a free advanced search function to offer improved reasoning skills and seamless tool integration to provide expert-level responses.
Why It Matters: The launch of Baidu’s next-generation AI model comes at a critical juncture as Chinese companies strive to create advanced AI models to rival OpenAI and other U.S.-based firms. On Tuesday, at the World Governments Summit in Dubai, Baidu CEO Robin Li stressed the importance of persistent investment in data centers and cloud infrastructure, despite the cost-effectiveness of AI models like DeepSeek.
He emphasized the potential for cost reductions in foundation models, stating, “The inference cost [of foundation models] basically can be reduced by more than 90% over 12 months,” stated Li.
The CEO highlighted the need for more investment to develop smarter models, a statement made in the context of DeepSeek’s recognition for creating language models that perform comparably to top systems like OpenAI's GPT, but require significantly less computing power.
On Tuesday, Barclays maintained its ‘Equalweight’ rating on Baidu stock but raised the stock price target to $95 from $83 as a part of a re-rating of the broader Chinese ADRs.
The launch of DeepSeek’s revolutionary R1 model rattled the U.S. stock market in late January. NASDAQ lost 3.1%, indicating its highest one-day percentage decline since Dec 18, while chip giant, Nvidia NVDA lost close to $600 billion in market capitalization.
Image via Shutterstock
Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors.
© 2025 Benzinga.com. Benzinga does not provide investment advice. All rights reserved.
Trade confidently with insights and alerts from analyst ratings, free reports and breaking news that affects the stocks you care about.
