China has applied for way more patents in the sphere of generative AI than another country, the United Nations Intellectual Property Agency announced on Wednesday. The United States is in second place by a transparent margin.
While the technology offers the potential to extend efficiency and speed up scientific discovery, it also raises concerns about jobs and staff. According to the World Intellectual Property Organization (WIP), it was involved in around 54,000 inventions in the last decade to 2023.
More than 1 / 4 of those inventions got here about last yr – evidence of the explosive growth and interest within the technology since generative AI entered the mainstream public consciousness in late 2022, WIPO said.
The recent patent report, the primary of its kind, goals to trace patent applications as a possible indication of trends in artificial intelligence. It focuses only on generative AI and excludes artificial intelligence within the broader sense, which incorporates technologies resembling facial recognition or autonomous driving.
“WIPO hopes to give everyone a better understanding of where this rapidly evolving technology is being developed and where it is headed,” WIPO Director General Daren Tang told reporters.
In the last decade since 2014, over 38,200 generative AI inventions got here from China. That is six times greater than the US, where there have been almost 6,300. This was followed by South Korea with 4,155, Japan with over 3,400 and India with 1,350.
GenAI helps users create text, images, music, computer code and other content using tools resembling OpenAI’s ChatGPT, Google’s Gemini and China’s Baidu’s Ernie. The technology is utilized in many industries, including life sciences, manufacturing, transportation, security and telecommunications.
Some critics fear that GenAI could replace staff in certain professions or unlawfully use human-created content without fairly or adequately compensating the people behind it.
As with other sorts of patent applications, WIPO officials admit that the amount of GenAI patents is just not an indicator of their quality. This early within the technology, it’s hard to say which patents can have market value or transform society.
“Let’s see how the data and developments evolve over time,” Tang said.
The US and China are sometimes seen as rivals in the event of artificial intelligence. But in some ways, it’s US technology corporations which are leading the best way in developing the world’s most advanced AI systems.
“Looking at patents only tells part of the story,” says Nestor Maslej, principal investigator at Stanford University’s Institute for Human-Centered Artificial Intelligence. He adds that patent approval rates can vary depending on a rustic’s laws.
“When you look at the dynamics of AI, the important question is who is coming out with the best models, where are those models coming from, and at least by that metric, the United States seems to be really far ahead,” says Maslej, editor of Stanford University’s annual AI Index, which measures the cutting-edge.
According to this yr’s AI Index, 61 notable machine learning models got here from U.S. institutions in 2023, surpassing the 21 models from the European Union and the 15 from China. Among EU countries, France had probably the most models, with eight.
Another assessment says the US also has probably the most so-called foundational AI models – resembling OpenAI’s GPT-4, Anthropic’s Claude 3, Meta’s Gemini and Llama, that are huge, versatile and trained on huge data sets.
The US also leads China in private AI investments and the variety of newly founded AI start-ups, while China leads in industrial robotics.
