SenseTime chief scientist Lin Dahua targets cost efficiency in AI race

The competitive landscape of China's artificial intelligence sector continues to intensify as established players and newcomers alike battle for market dominance.

SenseTime, a veteran of the industry founded in Hong Kong in 2014, has successfully pivoted its operations to embrace the generative AI era.

During a recent discussion at the company's Science Park offices, cofounder and chief scientist Lin Dahua detailed how the firm is navigating steep research costs and hardware expenses.

A central pillar of this new strategy is the SenseNova U1 model, which integrates language and vision processing.

Lin Dahua emphasised that while international competitors like OpenAI's GPT Image 2 and Gemini's Nano Banana may offer "exquisite and beautiful" results, the SenseNova U1 is significantly more affordable.

"You may not need the top model in many cases when it can handle most tasks," Lin told CNBC. He further noted that "there is still a gap between us and the international frontier models... but our cost is much lower – it's very efficient."

The company is increasingly targeting enterprise clients who value reliability and lower price points. This focus on sustainability appears to be yielding results, as SenseTime narrowed its net loss by 58.6% last year.

While the firm has previously faced U.S. sanctions over surveillance allegations, which it denies, it is now looking toward growth in Southeast Asia and the Middle East.

Despite historical challenges, SenseTime remains committed to infrastructure investment, having recently integrated its audio functions with ByteDance’s Seedance technology to enhance its short-video tool, Seko.