USPTO: no room for artificial inventors

Not surprisingly, also the USPTO came to that conclusion: inventorship is limited to natural persons. Thus, this is in line with the EPO‘s and the UKIPO‘s recent decisions.

Similarly to the European cases, the decision issued on 22 April 2020 (here) came in response to two patent applications on inventions created by an AI system called “Dabus”, in the context of the Artificial Inventor Project. Dabus is also known as the Creativity machine, which was developed by Dr. Stephen Thaler, who is named as the applicant and assignee in the patent applications. The Artificial Inventor Project has filed patent applications via the Patent Cooperation Treaty in various countries including the US, UK, Germany, and China.

The applicant (Dr. Thaler) referred that the Creativity machine is programmed as a series of neural networks trained with general information to independently create. It was the machine, not a person, who recognized the novelty and salience of the inventions at stake.

The application was listing a single inventor (Dabus) and the family name “invention generated by artificial intelligence”.

The main argument of the USPTO is similar to that of the EPO:  US patent law refers to inventors as humans, individuals, or persons. The term “inventor” therefore means the individual who invented the subject matter of the invention. The patent statues preclude a broad interpretation where “inventor” could be construed to cover machines. This view was confirmed by the Federal Circuit that (albeit referring to inventorship in the context of corporations) explained that patent laws require the inventor to be a natural person (see University of Utah v. Max-Planck-Gesellschaft zur Forderung der Wiessenschaften e.V., 734 F.3d 1315 (Fed. Cir. 2013), here: “[t]o perform this mental act, inventors must be natural persons“).

Also, the Manual of Patent Examining Procedure explains that inventorship requires “conception”. And “conception” is defined as a mental act, that is the formation in the mind of the inventor of the idea of the invention. Reference to “mental” and “mind”, again, points to a natural person.

The Office has also explained that inventorship has long been a condition for patentability, as naming an incorrect inventor is a grounds for rejection.

Last, the Office has refused to enter into any policy considerations on the advantages of supporting allowing AI as inventor, as in any case “they do not overcome the plain language of the patent laws”.

Is the end of the story? As the UKIPO said, further debate is needed. It will be interesting to see what local patent offices in other countries will say, especially after a Chinese court held AI-written articles protected by copyright – see here – although a human element still appears necessary.

Francesco Banterle

A requiem for torpedo actions? A catalogue of the most recent decisions on the issue

 

Actions

(i) relating to a European Patent

(ii) aimed at obtaining a declaration of non-infringement of different national portions of said European Patent

(iii) brought before Italian Courts

(iv) against a patentee who is not domiciled in Italy

are commonly known as Italian torpedoes (the expression is a fortunate coinage of Mario Franzosi).

These actions are often started to take advantage of the rules concerning lis pendens. Where a non-infringement action is pending before a Court (say an Italian Court) with regard to the national portion of a European Patent (say the German portion), any infringement action subsequently brought before the Court of another country (say a German Court) with regard to the same portion of the same European Patent may be stayed. This stay – and the consequent delay in ascertainment of infringement – is the goal of torpedo-launchers.

Of course, the slower the jurisdiction, the longer the stay. Thus, launching a torpedo is particularly effective in slow jurisdictions, like – at least in the past – that of Italy.

Traditionally Italian Courts have not looked with favor on such actions, seeing them as tools for misusing the (flaws of the Italian) legal system.

This traditional view was authoritatively expressed by a 2003 decision of the Joint Divisions of the Court of Cassation (the Windmöller decision), which deracinated torpedo actions on the grounds that Article 5.3 of the Brussels Convention of 1968 does not apply to non-infringement claims (Italian Court of Cassation, decision no. 19550/2003, in Giur. Ann. Dir. Ind., 2004, pp. 61 ff.). If it is the plaintiff itself which denies the existence of a harmful event, said the Court, then by definition there cannot be jurisdiction under Article 5.3 of the Brussels Convention.

A few years later, in the GAT case (here), the EU Court of Justice struck another blow against torpedoes by stating that the rule of exclusive jurisdiction laid down by Article 16.4 of the Brussels Convention of 1968 [corresponding to Article 22.4 of Regulation 44/2001] concerns all proceedings relating to the registration or validity of a patent

“irrespective of whether the issue is raised by way of an action or a plea in objection” (the principle was later incorporated in Article 24.4 of Regulation 1215/12).

As a result – and even regardless of the restrictive interpretation of Article 5.3 of the Brussels Convention given by the Windmöller ruling -, in order not to fall into the exclusive jurisdiction under Article 16.4 of the Brussels Convention and 22.4 of Regulation 44/2001, torpedo actions had to be based on pure non-infringement arguments.

In this scenario, most certainly not favorable to torpedoes, a decision of the Joint Divisions of the Court of Cassation reopened the discussion on the admissibility of this kind of action (Court of Cassation, 10 June 2013, in Giur. ann. dir. ind., 2013, pp. 60 ff.: the Asclepion case; a full English translation of the decision has been published in IIC, 2014, pp. 822 ff.). In the context of a non-infringement case brought before the Court of Rome by a German company with regard to the Italian and German portions of a European Patent owned by a US company, the Joint Divisions – quoting, word for word, the EUCJ Folien Fischer decision (here) – stated that

(aArticle 5.3 of Regulation 44/2001 … must be interpreted “as meaning that an action for a negative declaration seeking to establish the absence of liability in tort, delict, or quasi-delict falls within the scope of that provision”

and also that

 (btherefore [pertanto, in the original Italian text] with regard to the declaration of non-infringement sought … before the Court of Rome, the Italian Courts must be considered to have jurisdiction, on the grounds that they are the Courts of the place where the harmful event may occur, also with regard to the “German portion” of the European Patent” [my translation].

The 2013 Asclepion ruling certainly lent itself to interpretation as a general permit to launch torpedoes, at least when they do not rely on any arguments of invalidity. This is, indeed, what the Court said. But, looking more closely, a missing step can be seen in the reasoning of the Asclepion ruling, between statements (a) and (b). This significantly limits its practical impact.

If closing the jurisdictional door of Article 5.3 of Regulation 44/2001 to non-infringement actions, as the Windmöller 2003 decision did, fatally blocks any attempt to launch torpedoes, interpreting Article 5.3 as covering both infringement and non-infringement actions does not necessarily mean – in spite of the adverb “therefore” used by the Court of Cassation – that the Italian Courts have jurisdiction under this provision over actions concerning the foreign portions of a European Patent. To this end, something more must exist, and namely a link with Italy.

It is precisely because of the (non-) existence of this link that the torpedo actions brought before Italian Courts after the Asclepion ruling have been stopped.

In Schindler v. Otis, in line with Asclepion, the Court of Milan (Judge Dr. Marina Tavassi; decision of 27 January 2014, published in Giur. ann. dir. Ind., 2014, pp. 741 ff.) admitted that, in itself, a non-infringement action may be brought against a foreign defendant under Article 5.3. Nevertheless, it declined its jurisdiction on the non-infringement claims submitted against a US patentee, insofar as the Spanish portions of the European patent in suit were concerned. According to the Court,

Italy can be the place where the harmful event occurs or may occur only with regard to the Italian portion of a European patent, because there cannot be harm – actual or potential – caused in Italy by the alleged infringement of the Spanish portions of the patents in suit. Indeed, any national portion of a European patent is effective only in the relevant country and may be infringed only in that country” [my translation].

Therefore, in the opinion of the Court of Milan, when it comes to the infringement/non infringement of foreign portions of European patents owned by foreign patentees, there can never be a link with the country of Italy.

In the more recent Basf v. Bayer case, the Court of Milan (Judge Dr. Alessandra Dal Moro; the decision, dated 14 December 2016, published in Riv. dir. ind., 2017, I, pp. 309 ff. and also available here) fully confirmed the above approach and declared its lack of jurisdiction over the non-infringement of the foreign portion of a European patent (in particular, the German portion). In doing so, the Court explicitly dealt with the reasoning of Asclepion, clarifying why, in its view, the Court of Cassation’s decision could not be read as allowing torpedoes:

the fact that article 5.3 of Regulation 44/2001 (now replaced by Regulation 1215/2012) may be applied to non-infringement actions in no way changes the stance of the Supreme Court with regard to the existence of jurisdiction in the case of infringement (or non-infringement) of non-Italian portions of European Patents. A claim seeking a declaration whereby certain conduct does not constitute infringement of a patent means that the patent is effective, and this effectiveness is geographically limited to the perimeter of the legal system to which it refers. Outside that perimeter of effectiveness, it is not possible to discuss infringing or non-infringing conduct and it is therefore not even possible to allege an, actual or potential, ‘harmful event’” [my translation].

A further decision was issued on torpedoes in 2014, but on different grounds. In Agilent v. Oerlikon, the Court of Genova (Judge Dr. Rossella Silvestri; the decision, dated 23 April 2014, is available here) declined its jurisdiction because the torpedo-launcher had raised invalidity arguments in support of non-infringement claims concerning the German portion of a European patent and a German utility model. The claims thus fell into the exclusive jurisdiction of the Courts of the place of registration.

Hence, all the torpedoes launched in Italy after the Asclepion ruling have been dismissed on the grounds of lack of jurisdiction. Does this mean that the torpedo saga has finally come to an end?

Riccardo Perotti

Time limits for amending patent claims in pending litigations: the opinion of the Tribunal of Milan

The Tribunal of Milan (full text here) has recently ruled (for the first time) over a thorny issue concerning the Patent Limitation Procedure regulated by art 79.3 of the Italian Industrial Property Code (“CPI”). The norm states that “In a proceeding concerning nullity, the owner of the patent has the right to submit to the Court, at any stage or instance of the trial, modified claims that remain within the limits of the content of the patent application as initially filed and that does not extend the protection conferred by the patent granted” (emphasis added).

T1he facts of the dispute are simple. Thermorossi S.p.A. claimed that several competitors infringed its European patent’s Italian Portion No EP 2 083 221 on “heating apparatuses such as pellet-fired stoves and thermostoves”. As counterclaim, the defendants filed an action for the declaration of invalidity of EP ‘221.

Since an Opposition was filed against EP ‘221 and the EPO decision (through which the Patent was amended) was published only after the Italian final hearing, the owner presented to the Tribunal a set of amended claims only in the final statement.

Despite Thermorossi had known EPO’s decision only after its final hearing, Milan’s Tribunal stated that the new set of claims could not be considered in that procedure because the final hearing had already identified the object of the trial. According to the Tribunal, in light of the Constitutional principle of reasonable delay (Article 111.1 Italian Constitution), the right to file a new set of claims cannot be exercised after the final hearing.

Due to this reason, the Tribunal took into account the patent as originally approved by the EPO. Furthermore, it declared the Italian portion of EP ‘221 invalid and consequently rejected all the requests filed by Thermorossi.

2.jpg

The Tribunal rightly points out that Article 79.3 CPI should be interpreted in light of the Constitutional principle of reasonable delay. This consideration is certainly well founded considering that the limitation procedure (which requires the modified claim to be within the content limits of the patent application as initially filed and does not allow to extend the protection conferred by the patent previously granted), often requires an in-depth technical analysis. However, while it is true that in applying Article 79.3 CPI the Judge has to respect the principle of reasonable delay, it is equally true that the Court must make a case by case assessment. The words “at any stage of the trial ” seem to admit the reformulation of the claims even after the final hearing if this is deemed necessary for the overall economy of the judicial procedure (the opposite solution would require the institution of a new trial).

Giovanna Sverzellati

Tribunal of Milan, IP section, decision of 20 October 2016, No. 11544

Huawei v. ZTE: Enforcing standard-essential patents as abuse of dominance

Following the request for a preliminary ruling issued by the Landgericht of Düsseldorf (Germany), on July 16th 2015, the European Court of Justice addressed the question whether or not, and at what conditions, the firm holding a standard-essential patent (SEP: namely, a patent essential to produce manufactures in compliance with a particular standard), which has committed to grant a license to third parties on FRAND terms, abuses its dominant position by seeking injunctions against alleged infringers (decision available here).

The long-awaited judgment of the Court confirms the general approach adopted by the Commission in both Motorola and Samsung cases (available here and here). However, while the Commission had merely stated that enforcing a SEP in court can constitute an abuse of dominance under certain circumstances, the ECJ decision goes further, clarifying what those circumstances are. In particular, the Court held that seeking an injunction against an alleged infringer does not violate competition law when the following conditions are met:

  1. prior to bringing the action, the SEP holder has informed the alleged infringer of the violation of its intellectual property right, specifying the mode of infringement;
  2. the SEP holder has presented a written offer for a license on FRAND terms to the “infringer” (which has previously expressed its intention to conclude a license agreement). The offer must include all the relevant conditions of the agreement, in particular the royalty rate applied and the way it is calculated;
  3. the potential licensee has not “diligently responded to [patent owner’s] offer in accordance with recognized commercial practices in the field and in good faith”, and has continued to use the protected technology.

The decision of the Court of Justice seems to subordinate the finding of abuse to the “bad faith” of both SEP owners and producers of standard-based products. On the one hand, the formers have to concretely fulfill the obligation to the standardization bodies consisting in giving a license on FRAND conditions to third parties. Indeed, a patent cannot obtain the SEP status unless the legitimate holder undertakes to grant a FRAND license to anyone who may require it, in order to prevent the SEP holder from “reserv[ing] to itself the manufacture of the products in question”. Thus, it is not surprising that, according to the ECJ, the patent owner may incur in an abuse of dominance if it seeks an infringement injunction without even submitting a licensing agreement to the alleged infringer.

On the other hand, the ECJ imposes the obligation of good faith also to the potential licensee, which – having decided not to accept the offer submitted by the patent owner – “may rely on the abusive nature of an action for a prohibitory injunction or for the recall of products only if it has submitted to the proprietor of the SEP in question, promptly and in writing, a specific counter-offer that corresponds to FRAND terms”.

The ECJ judgment has definitely the merit of striking a reasonable balance between the interests at stake: those of the potential (and willing) licensees, which supposedly made specific investments relying on the FRAND license promised by the SEP holder; and those of the SEP holder itself, which should be granted the right to effectively protect its intellectual property rights from free-riders.

Nonetheless, the intervention of European judges leaves some issues unsolved. Firstly, the decision does not explain when a license can be considered FRAND. Secondly, it does not answer the question whether holding a SEP implies, per se, a dominant position on the market (actually, the referring court had not asked about the finding of dominance). At first glance, there seems to be the glimmer of an opening for such a conclusion. A passage of the sentence, in fact, reads as follows: “[…] the patent at issue is essential to a standard established by a standardization body, rendering its use indispensable to all competitors which envisage manufacturing products that comply with the standard to which it is linked. That features distinguishes SEPs from patents that are non essential to a standard and which normally allow third parties to manufacture competing products without recourse to the patent concerned and without compromising the essential functions of the product in question” (emphasis added). This appears quite close to an irrebuttable presumption of dominance. A debatable position: in my view, the presumption should be rebuttable (as suggested by the Advocate General Whatalet in his opinion, available here), hence the judges should continue assessing, case-by-case, a situation of actual, effective dominance.

Piera Francesca Piserà

CJEU, 16 July 2015, case C-170/2013, Huawei Technologies Co. Ltd. v. ZTE Deutschland GmbH.

Product-by-process patent = product patent??

Worthy of comment is a recent decision of EPO’s Enlarged Board of Appeal (March 25, 2015) in the cases Tomato II and Broccoli II (respectively, Case Number G2/12 here and Case Number G2/13 here). Both cases relate to the patentability of plants and plant matter under Article 53(b) EPC. The Tomato II case concerns a “method for breeding tomatoes having reduced water content and product of the method”, while the Broccoli II case concerns a “method for selective increase of the anticarcinogenic glucosinolates in brassica species”.

The EPO’s Enlarged Board of Appeal stated that even a product-by-process patent is a “product patent” autonomously protected from the process that describes the former and from which the implementation as product depends.

The EPO’s approach (see, in each decision: Reasons, Legal erosion of the exceptions to patentability,§ 6.(b), p.61) is open to criticism. On the one side it risks paralyzing the competitive dynamics of process innovation, in contrast an interpretation of articles 64(2) EPC and 34 TRIPs whereby third parties are allowed to make the same product with a totally different process—hence,   without any absolute block imposed by pre-existing product patents that are the fruit of a totally different intellectual process. On the other side, EPO’s opinion ‘forgets’ that in absence of that process, the product wouldn’t even exist, as the facts of the cases highlighted that no other processes were available to achieve the same result. The final outcome of said decisions is even more debatable as the process through which the product was described and claimed was essentially biological , thus as such non patentable: with the paradoxical ultimate result of allowing the patenting of a product that could be realized only through a process not patentable by definition!

Gustavo Ghidini

Gianluca De Cristofaro

EPO, Enlarged Board of Appeal, 25 March 2015, Case Number G2/12 and Case Number G 2/13, Tomato II and Broccoli II

updated on 12 November 2015

ECJ: the “supplementary protection certificate” for protein D in Synflorix. A tough interpretation of Regulation (EC) No.469/2009

On January 15, 2015, the Court of Justice of the European Union (ECJ), in Arne Forsgren v. Österreichisches Patentamt C-631/13, issued a decision regarding the granting of a Supplementary Protection Certificate (SPC), requested by Arne Forsgren, for Protein D used in Synflorix, a vaccine authorized in European Union to protect from diseases caused by the bacterium S.Pneumoniae (full text here). As per the Regulation (EC) No.469/2009, the SPC covers the cost of research leading to the discovery of new “products” in order to ensure sufficient protection to encourage pharmaceutical research, which plays a decisive role in the continuing improvement in public health. In the present case, the Protein D contained in Synflorix is bond to some polysaccharides as a “carrier” and not associated to any pharmacological effect on its own. Based on the concepts provided in Regulation (EC) No.469/2009, can a SPC be attributed to the Protein D, even if this substance is not considered an active ingredient? The commission rejected the request, based on the interpretation of Regulation (EC) No.469/2009 article 1, lett.b and article 3, lett.a and b. according to which the “carrier” protein could be protected by a SPC. Otherwise, in respect of the article 3, lett.b the SPC protection is granted only if Protein D may be categorised as an “active ingredient” and if this action is covered by the therapeutic indications of the marketing authorization. The decision to provide the SPC protection should be carefully taken: potential misapplications may result in risk to monopolize the market, viceversa the refusal to grant supplementary protection might take the interested company in a disadvantageous position in relation to market competitors.

Grazia Maria Gaspari

Court of Justice, 15 January 2015, C-631/13,  Arne Forsgren v. Österreichisches Patentamt