Patents in Crisis: Is there a Solution in Sight?
Remarks prepared for presentation to the Emily C. and John E. Hansen IP Institute at Fordham Law School on September 19, 2019

I. Introduction
II. The Four Sledgehammers Affecting Confidence in U.S. Patent System
III. Historical Context
IV. How will Confidence be Restored in Our System
A. Standards for Obviousness
B. Injunctions for Patent Infringement
C. Patent Eligibility
D. Third Party Challenges to Patent Validity in the Patent Office
V. Internationally
A. Standard Essential Patents
B. “Follow-On” Pharmaceutical Patents
C. Artificial Intelligence
VI. Conclusion

I. Introduction

Is there a patent crisis? It depends on whom you ask. Neither Europe nor China seem to see more than relatively minor difficulties in particular areas, although there are some issues looming in patent-heavy sectors such as telecommunications and pharmaceuticals. In the United States however, we are still wrestling with the consequences of using four sledgehammers to crack a fairly small nut resulting from the dot com boom and bust.

Traditionally, patents were seen as shields providing a means to protect an investment, and patents were valued on corporate balance sheets at cost. Then accountants started trying to attach a “real” value to such assets making the acquisition of patents more desirable, but also creating a need to justify the values that were being attributed to them. This tendency coincided with an increased willingness of the courts to recognize certain types of inventions as patentable that had hitherto been regarded as unpatentable following the 1983 case of Paine Webber v. Merrill Lynch, where Paine Webber failed in an attempt to have a patent relating to Merrill Lynch’s cash management accounts declared invalid. The patent claimed a system for processing and supervising a plurality of subscriber accounts by the interaction of various means.

This trend reached a high point in the 1998 case of State Street Bank v. Signature Financial where the Federal Circuit Court of Appeals upheld a patent relating to the allocation of costs between brokerage accounts on the basis that such allocation was a “useful, concrete and tangible” result. Arguments that all that these patents claimed was effectively just a computerization of old techniques that were rejected on the basis that computerization itself rendered the technique patentable. Consequently, coinciding with the dot com boom (indicated by Wikipedia to run from 1994 – 2000), a large number of patents came to be granted on what we call applications or “apps”.

Then came the dot com bust (2000 – 2004). Those who acquired the companies that had failed in the bust found themselves as owners of large numbers of patents but not much in the way of tangible assets. This led directly or indirectly to the need to use patents as swords as well as shields since the only ways to obtain a return on the investment was either to license the patents or to obtain damages by enforcing them. This led to increased scrutiny of the way in which patents were obtained and enforced, particularly after the National Academy of Sciences and other national institutions issued a report in 2004 that the quality of patents issued by the United States Patent and Trademark Office needed to be improved.

II. The Four Sledgehammers Affecting Confidence in the U.S. Patent System

This in turn led to the use of the four sledgehammers referred to above by virtue of:

1. The Supreme Court’s decision in KSR v. Teleflex changing the standards used to determine whether an alleged invention was obvious;
2. The Supreme Court emphasizing in e-Bay v. MercExchange that the patent statute requires consideration of equitable principles in determining whether an injunction should be granted once a patent had been found to be valid and infringed;
3. The Supreme Court’s revision of the law on what constitutes a patent eligible invention in Bilski, Mayo, Myriad and Alice; and
4. Introduction by statute of new procedures whereby third parties could challenge the validity of patents in the patent office even though they were not engaged in any case or controversy with the patent owner.

All have proved to be extremely controversial. All have reduced confidence in the system.

III. Historical Context

This is not the first time that overreaction has damaged a patent system. The Netherlands abolished patents for a while in the nineteenth century; Bismarck was not an enthusiast and The Economist had been running anti-patent articles since the 1860s. However, The Economist may have been guilty of the same error as that of which it has recently accused Karl Marx when it opined that Marx had been correct in his analysis of the situation in Das Kapital, but wrong in his proposed solution in The Communist Manifesto because this discounted the importance of providing the right incentives to individuals. The Soviet experience bears out the importance of selecting the correct incentive. Under the Soviet system, inventors did receive rewards from the state, which encouraged scientific achievement, but few of these achievements resulted in products useful to the general public because there was no incentive to develop such products. What was needed was an incentive that would not just reward an inventor, but that would spur product development. As Adam Smith noted two hundred years ago:

If the legislature should appoint pecuniary rewards for the inventors of new machines etc. they would hardly ever be so precisely proportioned to the merit of the invention as (the grant of a fourteen year “monopoly”). For here, if the invention be good and such as is profitable to mankind, he will probably make a fortune by it; but if it be of no value he will also reap no benefit ¹.

The Chinese Communist Party has not repeated the Soviet mistake and after early hesitation has invested heavily in a patent system since it was first introduced in China in 1985. Today 8.6% of all U.S. patent applications originate from China through PCT filings and China is at the forefront of developing products having commercial value. China is the source of 43% of the world’s total patent applications.

Before looking at the four sledgehammers noted above, we should point out that an ambiguity has always existed as an undercurrent to the Constitutional mandate to provide exclusivity to promote the useful arts, namely whether such promotion is intended to provide an economic tool to help commercialization of useful inventions or whether the exclusivity is provided simply as a reward for a contribution of useful inventions to the stock of human knowledge. The patent systems in most of the rest of the world follow the former view. The Paris Convention has since 1883 provided that countries may, but are not required, to provide remedies for non-use of a patented invention. Article 1 of the Japanese Patent Law provides:

The purpose of this law shall be to encourage inventions by promoting their protection and utilization so as to contribute to the development of industry. (emphasis added).

In the United States by way of contrast, in 1908 in Continental Paper Bag Co. v. Eastern Paper Bag Co. the Supreme Court rejected the argument that the patent owner was not entitled to an injunction even though the patent had been found to be infringed because the patentee had not itself used the patented invention. The Supreme Court dismissed the argument that patent ownership without developing the invention failed to promote the useful arts, noting that the issue of non-use by a patent owner had been considered by Congress in 1836 and that Congress had decided against any diminution of a patent holders rights simply because the patentee had failed to use the invention itself.

IV. How then will confidence be restored in our system?

A. Standards for Obviousness

It was not until the middle of the twentieth century that patent statutes started to include specific requirements that in order to be patentable an invention not only had to be new, but also not obvious. Before then courts had reasoned that minor changes from what had been done before were not inventions and so were not entitled to patent protection. The United Kingdom adopted a statutory requirement in 1932 that to be patentable an invention could not be obvious and required an inventive step. In 1952, the United States added 35 U.S.C. 103 to the statute to require that the invention must not be obvious to a person having ordinary skill in the art. Perhaps because of the concern that a patent owner could sit on his rights as noted above, U.S. law came to focus on the question of whether issuance of a patent would take away a third party’s ability to carry out acts that were obvious from what had been done before for any reason. In contrast, many other countries have focused on the question of whether a patent claims an obvious solution to the specific problem addressed by the inventor. These different approaches sometimes produced different results, but until the KSR decision, anecdotal evidence seems to indicate that similar conclusions were normally reached whichever approach was used.

In the KSR decision, the Supreme Court effectively raised the standard for securing patent protection in the United States, to prevent patents from being granted for inventions that had too low a level of inventivity. As Justice Kennedy put it in a unanimous decision of the Supreme Court:

… as progress … is expected in the normal course [of events], the results of ordinary innovation are not the subject of exclusive rights under the patent laws. Were it otherwise, patents might stifle rather than promote the progress of the useful arts.

In its opinion the court emphasized that, even if the prior art did not include a teaching suggestion or motivation pointing to the claimed invention, the invention could still be regarded as obvious, noting.

Often, it will be necessary for a court to look to interrelated teachings of multiple patents; the effects of demands known to the design community or present in the marketplace; and the background knowledge possessed by a person having ordinary skill in the art, all in order to determine whether there was an apparent reason to combine the known elements in the fashion claimed by the patent at issue.

Furthermore, the court noted that when the prior art suggested a finite number of identified predictable solutions, a person of ordinary skill in the art has good reason to pursue known options within his or her technical grasp which would render the actual solution found obvious for patent purposes.

As a practical matter, the effect of the decision has been to make it more difficult to obtain protection and to resist challenges of invalidity outside the field of chemistry. In chemistry, the courts have tempered the KSR approach by requiring that, when considering whether a new compound was obvious, one has first to consider whether there was any reason to modify the compound from which it was derived and also that, when evaluating options open to the person of ordinary skill in the art, one has to consider whether any option would be considered as having a reasonable expectation of success in producing something useful. In other technical areas, however, securing protection has become more difficult, especially when the invention involves substitution of one known element for another to obtain predictable results. This essentially poses the test as whether the prior art teaches that a change from the prior art could be made rather than posing the test, as set out in European case law, as whether the prior art teaches that the change should be made.

As noted above, the approach taken in the United States differs from much of the rest of the world and can produce different results. Viewed from the perspective of the public, however, the approach has appeal.

B. No Automatic Injunctions for Patent Infringement

The e-Bay decision, although arguably simply applying the words of the statute as written in holding that injunctions against the infringement of patents should be based on equity, has been argued by many as flying in the face of the Constitutional provision giving Congress the power to grant exclusive rights to inventors who obtain patents for their inventions ². The idea that the courts have the power to modify that exclusivity is contrary to the idea that a patent is a private property right and more consistent with the view that it is an economic tool. This takes us back to the nineteenth century debates about whether the Constitution’s use of the word “secure” rights to an inventor, thereby implying recognition of the inventor having a pre-existing right in his or her invention, was intended as a clean break from the English position under which a patent was a grant by the state and the language of the statute provided that such grant could be withheld if it would be “generally inconvenient”. (See, for example Reyna J’s dissent in Cascades Projection v. Epson).

As a practical matter, the decision in E-Bay made it more difficult for those who had bought up patents following the dot com bust to obtain injunctions (for a while the rule of thumb was that one could only obtain an injunction against a direct competitor). However, it has also made it more difficult for research organizations to obtain injunctions since it is difficult for them to show that they cannot be compensated adequately by money damages, one of the equitable provisions that must now be considered before granting an injunction. There are proposals in Congress to modify the law to encourage the grant of injunctions as set out in the STRONGER PATENTS ACT on which hearings took place on September 11. These proposals have, however, attracted significant criticism particularly since there does not seem to have been any groundswell against recent decisions in which injunctions have been denied and damages in the form of a continuing royalty awarded in lieu.

My own view is that if there are problems with the way the system is working, it is probably better to leave flexibility to deal with such problems at the remedy end of the system than to try to provide workable statutory definitions limiting the grant of patents at the front end.

This brings us to the topic where the Supreme Court’s actions have been perceived to have caused the most damage.

C. Patent Eligibility

The United States Supreme Court has long held that inventions directed to abstract ideas, laws of nature and natural phenomena are not suitable for patent protection. In the past decade, however, it has applied these requirements in ways that many have found surprising and disturbing. The Supreme Court has given no clear guidance as to what it understands to be an abstract idea, which in itself has led to uncertainty and inconsistency in lower court decisions, an issue to which we will return. With respect to the other two categories, the Court has, however, made controversial and possibly scientifically erroneous decisions.

In the case of Myriad, the Supreme Court held that, despite years of practice to the contrary by the United States Patent and Trademark Office, patent protection could not be obtained for an isolated gene. This puts the United States at odds with much of the rest of the world where a gene that has been isolated from its genome (and so is different from how it occurs in nature) can be patented as long as the function of the gene has been determined and described in the patent.

In Prometheus v. Mayo, the Supreme Court held that the metabolization of a drug was a natural law so that determining the results of such metabolization as an aid to diagnosis was not eligible for patent protection because it claimed a natural law. This again is contrary to the position at least in Europe where the Enlarged Board of Appeals of the European Patent Office has construed a prohibition on methods of diagnosis as applied to the human or animal body in such a way as to allow claims to methods of diagnosis as long as the claims are written in such a way as to avoid reciting a positive act of performing an operation on the body. The Supreme Court’s decision seems scientifically suspect since there was nothing natural about giving the patient the drug in the first place and it is difficult to see how this is any more natural than any chemical reaction that takes place when chemicals are mixed together under conditions under which they will react. Nevertheless, in the recent case of Athena v. Mayo, judges of the Federal Circuit, while expressing concerns about the Supreme Court’s view of the science, were unwilling to consider en banc a holding that a claim to a diagnostic method lacked patent eligibility. There is a possibility that this case may be appealed to the Supreme Court.

A similar issue has already been appealed to the Supreme Court in Vanda v. West-Ward, where the Federal Circuit upheld the validity of a claim to a method of treatment where the inventive contribution was essentially in the diagnosis steps employed to determine the treatment to be given. In Mayo v. Prometheus, the Supreme Court had stated that it did not intend its decision to foreclose patenting of methods of treatment. In Vanda, the majority of the Federal circuit panel based their conclusion that the invention was patentable on that reasoning. Prost, CJ, dissented saying she could see no real difference from Prometheus because all that the patent claims did was add an obvious treatment step to what had been found unpatentable in Prometheus v. Mayo. The Supreme Court has not yet decided whether to grant certiorari in the Vanda case, having requested the solicitor general’s advice on the government’s view on the issue.

Anecdotal evidence indicates that research on methods of diagnosis has been moving from the United States to other countries, in particular Europe. It is difficult to see how limitations on biological inventions set out in Myriad and Mayo can be solved without either Congress “overruling” the Supreme Court or perhaps, in the case of diagnosis, the Supreme Court acknowledging that it got the science wrong in Prometheus v. Mayo and that placing a chemical in an unnatural environment is not really “natural”. Any attempt to resolve the question of patentability of genes and diagnosis by Congressional action is likely to prove difficult because it will inevitably become entangled with questions about the cost and most effective way to provide health care.

The other area where the Supreme Court’s action caused consternation has been slowly working itself out at the Federal Circuit level. Many of the early decisions following Bilski and Alice related to patents that were in effect directed to apps that had been patented during the dot com boom. In most cases, the patents were found to be invalid as being directed to an abstract idea and lacking any inventive way for implementing that idea. This led to a perception that most software patents were invalid. More recent case law has changed this perception in that the Federal Circuit has upheld the validity of patents relating to software improving the efficiency of a computer, improving the security of a computer, and using a computer to improve synchronization of lip movements of animated characters with their voice-overs. Based on such case law, the Patent Office has issued guidance to the effect that patent applications should be rejected for being directed to abstract ideas only if they are directed to:

A. Mathematical concepts – mathematical relationships, mathematical formulas or equations, mathematical calculations;
B. Certain methods of organizing human activity – fundamental economic principles or practices (including hedging, insurance, mitigating risk); commercial or legal interactions (including agreements in the form of contracts; legal obligations; advertising, marketing or sales activities or behaviors; business relations); managing personal behavior or relationships or interactions between people (including social activities, teaching, and following rules or instructions);
C. Mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion).
Furthermore, Patent Office guidance recognizes patent eligibility of practical applications of these ideas that go beyond saying simply “apply the concept” and that are restricted to inventive applications of the ideas. The guidance indicates that this occurs for example where the practical application results in:

(i) improvements to the functioning of a computer;
(ii) improvements to any other technology or technical field;
(iii) use of a particular machine, or
(iv) effecting a transformation or reduction of a particular article to a different state or thing

The Federal Circuit is, of course, not bound by Patent Office guidance, but since it seems reasonably aligned with specific decisions of the court, it may be hoped that the law will in fact settle along these lines.

There are proposals in Congress put forward by Senators Coons and Tillis to “restore” patent eligibility to a broader class of inventions than those listed above and to turn the clock back to the situation following the State Street decision or even earlier by abolishing judicial exceptions to patent eligibility and making anything that provides a specific and practical utility in any field of technology by human intervention eligible for patent protection. The economic situation and approach to obviousness has changed since State Street and so in the light of these changes could be viable. However, whenever Congress changes the law, the courts take a while to interpret the new language and I am fearful that where Congress intervenes, except where absolutely necessary, this leads to a further period of uncertainty. Furthermore, in other countries and regions, such as the EPO and Japan, where the legislature has tried to provide definitions of what ought not to be patented, the courts have had to step in and “interpret” such prohibitions so as to limit their effects when the practical consequences of limitations on, for example methods of treatment, became apparent. Any such list in the U.S. would likely hamper the courts in their attempt to reach the right conclusion and any over broad definition will bring us back to the debates of fifty years ago as to whether the Fosbury flop that won the high jump in the Mexico Olympics should have been patentable.

D. Third Party Challenges to Patent Validity in the Patent Office

Unlike most other countries, the United States traditionally provided no mechanism through which the validity of a patent could be challenged unless the challenger was involved in an actual case or controversy with the patent relating to the patent. Modest steps to change this situation started in the 1980s but were not very successful. It was almost inevitable therefore that when the clamor about “bad patents” climaxed in the 2000s, it was decided to follow the rest of the world and create procedures by which any third party could challenge the validity of a granted patent in proceedings before the Patent Office.

Unfortunately, when the America Invents Act (AIA) introduced such “opposition” proceedings in the form of inter partes review and post grant review, it indulged in overkill. While in most countries that have such proceedings, it is at least as difficult to have a patent held to be invalid in opposition proceedings before the Patent Office as it is when challenging validity before the courts in a defense to an infringement action, the AIA made it easier to succeed in a challenge in the Patent Office than in the courts.

A number of factors contributed to this:

1. The lower burden of proof applying to the party challenging validity in the review proceedings (balance of the probabilities) is lower than that in court proceedings (clear and convincing evidence).
2. For proceedings based on petitions filed prior to November 13, 2018, the fact that in the Patent Office Review proceedings the claims are given their broadest reasonable interpretation, whereas in district court proceedings a narrower interpretation aimed at preserving validity may be employed.
3. Until recent changes, Patent Office rules making it difficult for the patent owner to amend claims to restrict them to what was valid if the challenger showed that parts of what was claimed lacked novelty or were obvious.
4. The speed with which the review proceedings must be carried out, which for most cases is set by the statute as being no more than one year from the institution of proceedings.

These factors are only partly counterbalanced by the fact that, if one fails in a Patent Office review, one is estopped from later challenging validity in any district court or ITC proceedings on any ground that could reasonably have been raised in the Patent Office review proceeding.

As noted, some of these factors have already been addressed by changes in the rules to make the proceedings fairer to patent owners. If enacted, the STRONGER PATENTS ACT would make further changes by providing that the same burden of proof (clear and convincing evidence) should apply when challenging validity in these proceedings as in court and requiring that only those who had been threatened with patent infringement should have standing to bring such proceedings.

The first of these further proposed changes seems to me to have merit. There is no logical reason why it should be easier to have a patent declared invalid in relatively simple and cheap proceedings in the Patent Office than in court proceedings where matters can be investigated more fully. One of the objectives of the AIA was to clean out clearly invalid patents. The proposed change would not affect this.

The other proposed change is more problematic since it could preclude those who are thinking of developing a new line of business from being able to challenge patents that they perceive might in the future be obstacles. Furthermore, it would prevent public interest groups from making such challenges. Possibly one way to address this issue would be to have special standing requirements to bring such challenges that, as is the case with some types of opposition procedures in India, require that the challenger shows that it has “an interest” going beyond simple “public do-gooding” in determining whether the patent is valid or not.

V. Internationally

Turning to the International arena, two changes in the technical environment have presented problems that are currently being grappled with and one is positioned just over the horizon, but its effects are already being anticipated. These are, however, normally regarded challenges that can be resolved rather than crisis-generating problems.

The two changes in the technical environment that are already with us are:

1. the need for devices to communicate with each other and hence the need to consider whether any special provisions should apply to patents relating to such communication;
2. working out what is required to obtain and enforce patent protection for “follow-on” inventions in the pharmaceutical field.

Just over the horizon lies the question of how the patent system will handle inventions made by artificial intelligence.

A. Standard Essential Patents

The usefulness of an Apple iPhone would be considerably reduced if its users were unable to talk to someone whose cell phone was a Samsung Galaxy. To deal with this problem manufacturers of telecommunications and other equipment where interoperability is required tend to cooperate with each other to agree on standards that will permit such communication. Such cooperation may be through formal bodies such as the American National Standards Institute (ANSI) and the European Telecommunications Standards Institute (ETSI) or ad hoc agreements between those active in the field. There seems to be general agreement that companies participating in establishing a standard should not obtain an unfair advantage deriving from their ownership of patents (known as Standard Essential Patents – SEPs) that are essential to compliance with the standard that they have helped to set. Typically this issue is dealt with by the parties owning such patents agreeing that others who wish to operate according to the standard will be entitled to a license on a “Fair, Reasonable and Non-Discriminatory” (FRAND) basis, which sounds ideal in theory, but creates a number of real life problems:

1. Who determines the rate if the prospective licensor and licensee cannot agree;
2. Does a requirement to be non-discriminatory mean that the same rate applies to everyone or can other factors, such as the licensees own contribution to the standard be taken into account;
3. Can a patent owner sue for infringement of its patents and possibly obtain an injunction if it has offered a prospective licensee an agreement that it views as FRAND but the prospective licensee views as being too expensive;
4. Can a standards-setting body require that to participate in standards setting a company must agree that all related patents must be subject to a FRAND obligation or can a company hold back on committing to this obligation for patents that go beyond what is required to meet the standard but cover technology that gives improved performance while meeting the standard; and
5. What is the role of competition law as opposed to contract law in addressing issues arising with respect to SEPs? The US Federal Trade commission sees this as a competition law issue, as recently illustrated by its suit against Qualcomm. The Department of Justice see this as an issue of contract law. But what happens if the contract is, for example under French law (as is the case with most ETSI contracts) and the law does not recognize the concept of promissory estoppel and deals with third party beneficiaries in different ways from the United States.

All of these issues are slowly being resolved by the courts, but despite some differences in approach (for example whether courts in one country should try to impose global frameworks on how to determine whether a royalty rate is FRAND), there seems to be little doubt that we are working towards acceptable solutions to the issues outlined above and that the patent system will fulfil its objectives of promoting innovation.

B. The “Follow-on” Pharmaceutical Issue

Obtaining additional patents to the original patent for a new chemical compound is a bugbear for the anti-patent lobby who see such patents as ways in which Big Pharma obtains unjustified extensions of the periods of protection they obtain for their drugs. However, not all “follow-on” patents are the same.

1. New uses of old drugs represent a substantial part of modern drug development. Typically, an old drug has already been shown to be safe and so if it is effective for treating a new condition it can probably be brought to market more quickly than a new chemical entity, but it still requires substantial testing for efficacy before the FDA or other drug regulatory authorities will grant marketing approval. Patent protection is therefore needed to protect the investment required. A real issue arises, however, as to how to prevent the grant of patents for mere guesses at new uses for old drugs since such patents could prevent others who have actually investigated the properties of the uses in question being able to develop old drugs for such uses. In East Asia, this problem is avoided by requiring applicants to include data in their applications that point towards the prospect of success for such a use. Such data will normally also meet the requirements in the United States and Europe, but in these jurisdictions there is also the possibility of securing protection by including in the application an explanation as to why the compound would be useful for the proposed indication.
   What is emerging as a possibly more significant problem, particularly outside the United States, is how to prove infringement. In the United States, it is possible to obtain a patent for an actual method of treatment, specifying the condition being treated and the dosing regimen required. It is an act of infringement to request the FDA to grant marketing approval for such a use unless one certifies that one will not start marketing until the relevant patents have expired. This is not the case in the rest of the world where convoluted claim forms have evolved to protect this type of invention and there is no infringement until a product actually comes on to the market. The problem then is how do you prove that it is being sold for the patented use and what objective facts must be proved and whether the subjective intent of the supplier is relevant. There is as yet no uniform answer to these questions, but the courts are working on them. Based on what has been done on the question of how to provide patent protection for this type of invention in the first place, I am reasonably confident that a rational solution will be achieved, notwithstanding a 2-2-1 split in the United Kingdom Supreme Court on this issue last year.
2. No such problems of proving infringement arise with the other main type of “follow-on pharmaceutical patent”. Such patents are directed to specific formulation, salt forms or crystal forms of the original compound which are claimed to have unexpected properties that make them patentable. In some cases, this is true. Devising a composition that has greatly improved stability or greatly improved bioavailability provides significant medical benefits and, subject to the comments made above about granting patents for guesses, meets the traditional criteria for patentability. There are, however, legitimate concerns that claims of this type are sometimes made without sufficient justification so that “tough” examination of this type of invention may be justified.

There is a question, however, as to why pharmaceutical companies need to indulge in such “life cycle management”. Drug development remains expensive. There is no evidence that major innovative pharmaceutical companies make unreasonable profits as compared to other industrial companies – if they did everyone would be buying Pfizer, Merck and GSK stock rather than Apple and Amazon. If the normal period of patent protection does not result in sufficient profit to fund new research, some more transparent way of achieving this may be appropriate.

C. Artificial Intelligence

To date, artificial intelligence has largely been carrying out tasks that we humans could do for ourselves if only our brains were big enough, including:

1. data crunching to find hitherto unexpected correlations; and
2. “playing through” every possible way of applying a set of rules to determine the optimum outcome or some combination of the two as is used to “write” some of the music produced by artificial intelligence. Putting aside the question of who should be named as an inventor for innovations resulting from use of these techniques, the patenting issues are in many ways similar to those that arose when computer-implemented inventions first came on the scene.

As noted above, we ultimately came to the conclusion that improvement in the way the computer tool works are patentable as are inventive ways of using the computer tool in methods that otherwise meet the criteria for patent eligibility were patentable, whereas simply computerizing something that had previously been done manually or that was patent ineligible, did not meet the criteria for patentability. I suspect that we will reach the same conclusions for inventions relating to the currently available forms of artificial intelligence.

The problem arises when artificial intelligence takes the next step and starts re-writing the rules. What criteria are we to use to determine whether an invention is obvious if say, we ask artificial intelligence to design a version of chess that is more exciting as a spectator sport and it decides that the way to do this is to allow players to make three moves at a time rather than one move at a time? To that, I have no answer.

VI. Conclusion

Outside the United States, few have doubts about the value of the patent system as is shown by the increase in numbers of filings each year. There are issues of difficulty in some areas, but the predominant view outside the U.S. that patents are an economic tool is likely to enable the issues to be resolved. It is strange that the country which at present has greatest doubts about the system is the one which has traditionally been its strongest advocate. In my view, this is because we tried to deal with a real, but relatively minor problem following the dot com boom-bust cycle in four uncoordinated ways thereby resulting in overkill. As discussed, many of the consequences of this have been or are in the process of being mitigated. The one big issue that still presents a formidable challenge is what protection should be given for diagnostic methods and genes, which raise questions outside the patent field and perhaps need more broadly based consideration to reach an answer.