What impact will AI have on IP systems?

At its core, AI is a computer that has been programmed to mimic the natural intelligence of human beings by learning, reasoning and making decisions. Given the pace at which machine learning and AI is accelerating, there are questions that the patent system must answer: are current patent laws adequate for AI-related intellectual property? Are we ready for patents filed by AI? How can AI systems act as data analysis tools to enforce IP rights?

Present scenario

The answers to these questions are limited and inadequate, because current laws do not identify AI as an inventor, author or creator of intellectual property in any country. However, now that AI is able to produce poetry and artwork, generate 3D printing, write code, detect anomalies and develop inventions without any human involvement concerns about ownership have been raised. A Facebook report on the way in which chatbots converse with each other has given the world a glimpse into the future of AI. Researchers from the Facebook AI Research lab used a model that allowed the chatbots to converse freely. The chatbots eventually created and used their own non-human language and Facebook decided to shut down the project. The chatbots were apparently using machine learning to incrementally improve their conversational negotiation strategies as they chatted. This lone example tells us how little we know about the ways that AI thinks, learns and grows.

If AI is able to write unique code and algorithms, should the copyright be awarded to AI or the developer of AI? Without a patent, another developer could legitimately write another code in an AI software with the same functions of the prior developer of the copyrighted AI software.

A positive step

A major advantage of AI is its ability to provide repeated results as these systems are not hindered by inexperience or fatigue. In recent years, multiple AI-based systems have been built for IP professionals. Several companies have crafted systems designed to identify relevant prior art so that inventors and IP professionals can assess the patentability of their inventions easily. AI also provides a way to automate processes that were previously carried out manually. TurboPatent, a company that develops applications to automate and streamline the patent protection process, has introduce two AI products for patent lawyers. RoboReview, a cloud-based product that analyses draft patent applications and RapidResponse, a product that assists lawyers in writing responses to office actions. AI is also being integrated into translation services. Microsoft has developed its own translator app, which translates text, speech, images and street signs. Despite all of the positive things that AI offers, it still has substantial shortcomings – the most important of which is its lack of human reasoning. It is unable to carry out a ‘sanity check’ of results or inventions and lacks the personal experience that leads to a person’s intuitive response to situations. At the World AI Conference in Shanghai, Elon Musk described humanity as a “biological boot loader” for AI.

Comment

It will not be long before AI dramatically affects what it means to be human, which is both compelling and frightening. Specific IP legislation must be introduced, which will allow us to utilise, commercialise and protect the intellectual property that AI creates in an equitable fashion for inventors around the world.