According to the World Intellectual Property Organisation (WIPO) report 2019, approximately 3,40,000 patent families in AI and more than 1.6 million scientific papers related to AI were published from the year 1960 to 2018. The number has only risen as more and more enterprises are looking to leverage AI and ML technology. AI and ML play a vital role in many industries, and these industries seek ways to guard their innovations through copyright and IP protection.
The March 2021 CSET report on Mapping India’s AI potential highlighted that India had ten times as many AI-relevant patent applications in 2018 as compared to 2012. Patents are a barometer to measure innovation. According to the report, even though Indian owned AI patents are small compared to the research output, India ranks in the top 10 AI patent producing countries. The largest categories of AI patents in India are telecommunications, life sciences, personal devices and computing and business (making up 70% of India’s AI patents). The report concluded that India has the potential to position itself as an important player and a valuable partner in the AI ecosystem.
GlobalData, a data and analytics company, has found that innovations enabled by artificial intelligence take the top spot in the patent filings of 2020. “The huge number of patent filings in AI indicates a paradigm shift from theoretical research,” said Venkata Naveen, GlobalData.
The USPTO in 2019 granted 14,838 patents that mentioned AI or ML. Big Techs were unsurprisingly the top recipients of these patents, with IBM receiving 81 of these patents, followed by Microsoft at 56, Amazon at 51, Cisco and Facebook with 30 and Google at 26. As of 2020, IBM was the largest owner of active ML and AI patent families globally, owning 5,538 families. The company claimed the leading position from Microsoft in 2018, which now ranks third.
Since most AI-related innovations are enabled by data, the organisations that generate the most AI-related patents are often the ones that own the most data. This explains the particularly strong position of China as there are fewer obstacles to data collection in China compared to other countries. China has the best training data collections for speech recognition, human behaviour modelling and medical data, among others.
The incentive for companies to file for patents stems from the assurance of possessing the exclusive right to their discoveries. Innovation is an interactive process. Innovators are no longer isolated and self-reliant, but collaborate with upstream and downstream actors to benefit from increasingly complex knowledge flows. Tech innovation’s life cycle has different stages from invention, research, development and dissemination to market development and commercial diffusion. Patents can play a big role in the entire life cycle of AI innovations allowing competitive technologies to be protected and licensed to third parties in order to expand financial opportunities. Patent assurance is the reason that companies invest billions of dollars in research and development.
Apart from promoting innovation and economic growth, patents are also effective tools for technology transfer and knowledge sharing. A study found that 88% of US, European and Japanese businesses rely on patents to direct their own R&D and keep up with tech advances.
The culture of innovation thrives on a combination of patenting, publishing and contributing to open source and market experimentation.
Patents choke innovation?
Large companies have built their patent stockpiles as a purely defensive measure. The true potential of AI is difficult to comprehend unless all those patents see the light of the day. AI patent filings are a yardstick to measure a company’s future digital strategy. Patent hoarding leads to companies enforcing patent rights that are far beyond the patent’s actual contribution.
Companies are pursuing more draconian copyright laws in many fields in general and aggressive patent practices are undermining the very purpose of intellectual property. Moving even further, innovative startups become targets for acquisition fro major companies in an attempt to acquire expertise and portfolio.
Patents are often linked with innovation, and it is true that financial gain, exclusive rights and a stronger portfolio encourage patents, but patent rights do not last forever, and innovations eventually enter the public domain.