Artificial Intelligence and Patent Law: An 'Apparent' Slippery Slope
In 2020, researchers used a machine-learning system to help them build a potent antibacterial medication that works against a wide spectrum of microorganisms. AI is also utilized to aid in vaccine development, medicine development, content finding, space technologies, and ship design. Within the next several years, AI might be applied to various technologies. This is one of history's most severe challenges to patent systems.
Intellectual property law is predicated on the presumption that innovators are sapient; it presently grapples to cope with a machine inventor. Courts worldwide are grappling with this issue right now, as intellectual property applications citing an AI system as the originator have been filed in over 100 countries. Several organizations, including those in the United Kingdom, the United States, and Europe, are publicly discussing artificial intelligence legal issues.
The ramifications might be tremendous if tribunals and lawmakers find that AI-created ideas are not copyrighted. When the ROI is low, funders and businesses are far less likely to do useful research with AI pioneers. It's possible that society may lose out on crucial and possibly life-saving technologies.
Instead of forcing old patent laws to adjust to new technology, we propose that national governments adopt specialized AI patent laws to protect AI-generated concepts. Nations should also develop an international treaty to ensure that these rules follow standard standards and that any issues be resolved expeditiously. Researchers must be involved in both processes.
Not What, But Who
Artificial intelligence technologies that can remain competitive were not factored in by the authors of the world's first IP rights legislation, the Venetian Patent Penal code of 1474. They were also ignored in the 1883 Paris Declaration for the Preservation of Commercial Property, which established the global patent system. Even when the World Trade Organization finalized its Convention on Trade-Related Aspects of Intellectual Property Rights (TRIPS) in 1994, AI-generated concepts were virtually unknown. Today, the Treaties of 1883 and 1994 establish global patent regulations.
As per the TRIPS agreement, "all innovations, whether items or procedures, in all fields of innovation, provided that they are unique, imply an inventive step, and are feasible for industrial application." 'Inventions,' 'new,' 'inventive step,' and 'capable of industrial application' are all legal meanings of artistic terms. In essence, an item is not patentable if any of these factors are not met.
Because the difficulty is 'who' originated rather than 'what' was produced, AI-generated innovations provide a new challenge to the patent system. The biggest and most compelling issue that trademark registration agencies have had to deal with in dealing with such developments is whether or not the originator must be human. If not, one worry is that AIs will grow so pervasive that their ideas will overwhelm the patent system.
However, suppose artificial intelligence technologies become more informed and skillful than all other individuals in an area. In that case, it is unclear how a sentient patent applicant can decide if an AI's discovery was obvious. An AI system built to evaluate all available literature about a technology sector before creation would have a much larger library of knowledge than any human could. When weighed against all knowledge, almost everything becomes apparent. 4. If everyone had access to such AI technologies in the future, the 'inventive step' condition of patentability would be virtually impossible to achieve, and nothing would be eligible. A complete overhaul would be required.
The innovations are not noteworthy. It results from them. Thaler's international legal team, led by scholar Ryan Abbott of the Surrey University in Guildford, UK, began filing applications to patent offices worldwide in 2018, claiming DABUS as the inventor. These are believed to become the first instances of exploring to see if an AI system may be acknowledged as an intellectual under current laws. Legal practitioners and tribunals have had to adjudicate this issue and have begun to identify legal loopholes.
In the absence of precise laws detailing how to evaluate AI-generated discoveries, patent offices and courts are now required to interpret and enforce primary rules to the best of their abilities. This is far from an ideal condition. Governments should develop legislation that is specially tailored to AI inventiveness. We recommend three steps to accomplish this goal.
Listen and Understand
To begin, national governments and international agencies concerned with intellectual property policy (like the World Trade Organization) could perhaps conduct a thorough examination of the issues, facts, and perspectives. They should consult with stakeholders such as patent registration offices, professional bodies representing scientists and engineers, patient advocacy groups and consumers, commercialization bodies and business development, and IP legal specialists. Previous probes of this sort have resulted in modifications to several nations' intellectual property laws due to the emergence of the Web and the digital revolution. For example, over 2014 to 2018, the Australian government had public consultations on internet piracy, which resulted in rules that empower judges to limit access to websites that violate copyright.
Several governments, including the World Intellectual Property Organization and the European Union, have already initiated preliminary inquiries into AI-generated ideas. This is an excellent start, and patent systems elsewhere should follow suit.
These investigations must return to fundamentals to determine if safeguarding AI-generated ideas as intellectual property incentivizes the creation of valuable creations for civilization, as it is for other patented items. AI system developers may already receive protection of intellectual property rights through copyright in programming code and patents on the functioning of the software they create. Some individuals may wish to keep the result of AI-generated innovations in the public domains open for anyone to use for political or logical reasons. Others advocate for increased IP protection.
Changing current legal safeguards risks creating gaps. Thus, a more thorough law change is desirable. A perfect answer would be for nations to create a unique type of intellectual property law known as sui generis law. These custom-made rules are intended to protect sorts of creative work not covered by the 'big four' industrial designs, IP doctrines of copyright, patents, and trademarks. In some jurisdictions, they currently reward and safeguard investment in circuit design, new plant types, and databases.
Some may object to the proliferation of topic-specific variants of IP. On the other hand, a separate AI-IP concept has the benefit of being customized to the precise conditions under which AI creativity emerges. For example, if AI-IP is quicker and faster to produce, lawmakers may decide that it should be preserved for a shorter time than the typical 20-year duration of regular patents. This would motivate others to expand on innovations once the patent period expires. Moreover, while patents are normally granted to the inventor, policymakers may elect to divide the profits from an AI-generated innovation differently – maybe among the AI programmer, the person leading the AI, and the proprietor of the training dataset.
Countries that use AI-IP are more likely to attract R&D investment. On the other hand, royalties connected to the adoption of innovation may make it less accessible. The same thing is happening with pharmaceuticals and vaccines today: patents can attract investment that permits them to be produced, but people are losing out in nations that cannot afford the items or pay compensation to manufacture them. Meanwhile, some higher-income countries are providing fourth dosages. A balance must be struck while building AI-IP to prevent duplicating this type of injustice.
A country that frequently imports ideas may profit from not protecting AI-generated inventions. Instead of paying royalties, it might furnish its people with low-cost replicas of an expensive new medication produced elsewhere by an AI. However, it may also lose out on securing investment that will help expand the business.
This 'free-rider dilemma' is addressed through patent treaties. The TRIPS treaty was created partly to make it difficult for nations to bail out of providing the intellectual property insurance of patents. International treaties also control the use of innovations, trademarks, copyright, and other areas of intellectual property.
We believe that an international convention is also required for AI-generated inventions. It would provide consistent rules for protecting AI-generated ideas across several countries. This might be accomplished by drafting a new agreement or incorporating such provisions into an established international intellectual property accord.
Establishing such a pact would be a long-term, demanding goal. Critics may object to governments abandoning their ability to create domestic policy regarding AI-generated technologies, especially when the entire potential of AI is unknown. We believe that the worldwide advantages of an international treaty would be worth the effort and money spent negotiating it since it would prevent future uncertainty and disagreements.
A pact on AI-IP should also address a major flaw in the patent system. Patents are registered in each jurisdiction individually, and enforcement conflicts are usually decided by the nation's judicial system of registration. As a result, patent holders may face identical legal challenges in many countries. Because this method is cumbersome and expensive, some persons cannot manage to enforce or safeguard their patent rights.
On the other hand, an AI-IP pact may incorporate dispute-resolution procedures, possibly adjudicated by a specialized international court. Insight may be found in Europe's upcoming Universal Patent Court, set to begin operations next year, as well as different international arbitration tribunals.
Creating customized legislation and an international agreement will be difficult, but failing to do so will be far more difficult. AI is altering the way science is conducted and inventions are created. It must be purpose-fit to guarantee that IP legislation serves the public interest.
Knowledge Base Team
Knowledge Base Team
Knowledge Base Team
Knowledge Base Team
Explore More Special Offers
50,000 email package starts as low as USD 1.99, 120 short messages start at only USD 1.00