Skip to main content

Intellectual Property and AI part 2

8 min read
Read part 1
This is part two in a series of articles exploring recent developments in patenting artificial intelligence (AI) and machine learning (ML) based inventions. In part one, we examined some of the terminology used and the guidelines introduced by the European Patent Office. In this conclusion, we look at comparable advice from the US Patent and Trademark Office (USPTO) and some broad trends in patent filing in the last 20 years within this area of technology.

US Requirements


Eligibility criteria

The USPTO has been quieter on issuing explicit guidance for examining ML/AI patents than their European counterpart. As a starting point, some guidance can be attained from looking at the two criteria to determine whether the subject matter considered is eligible for a patent. This is not quite the same as the ‘patentability’ test (of ‘novelty’ and ‘non-obviousness’); instead it focusses on the form of the invention and some exceptions that are excluded from patent protection:

  • The first test is that the invention should fall into one of four categories:






Methods of manufacture


Compositions of matter

  • The second is that the invention be ‘useful’, and does not fall into any of the explicitly listed exceptions:


Laws of nature


Physical phenomena


Abstract ideas

For the purposes of ML and AI, the third exception (‘abstract ideas’) is the most relevant. Much has been made of this exception, including in the famous 2014 case Alice Corp. v. CLS Bank International1  (known as the ‘Alice’ case), which laid the groundwork of the modern USPTO interpretations of patentable subject matter for software patents.

An example of this ‘abstract idea’ exception in practice is highlighted in the case Purepredictive Inc v H2O AI Inc, where a California district court found that a pure-data application of ML was too abstract to qualify for patent protection.2  The exact wording found in the judgement was that the invention makes ‘use of computers only as tools, rather than provide a specific improvement on a computer-related technology’. The claims of the invention essentially comprise a three-step process of learning some mathematical functions, evaluating them, and selecting the best one. Since in this case (as in many others) the question of software patentability depends heavily on the ‘abstract ideas’ exception above, the USPTO released guidelines in January 2019 designed to clarify this exception specifically in the context of software inventions.3

Recent guidelines

The 2019 guidelines are not specifically focussed on AI or ML inventions. However, as with the EPO, a 2020 USPTO report on a 2019 ‘Request For Comment’ project stated that “A majority of commenters agreed that AI is viewed best as a subset of computer-implemented inventions. Therefore, this majority felt that current USPTO guidance…is equipped to handle advances in AI.”4   This may provide some insight into why there has been little specific AI and ML guidance, and inventors are left to use the more general software-patent guidelines.

The specific purpose of these 2019 guidelines was to clarify the concept of ‘abstract ideas’, dividing them into three categories:

  • Mathematical concepts
  • Methods of organising human activity
  • Mental processes

The USPTO has since given some specific examples of some fictitious inventions where (all else being equal) the question of patentability rests on these criteria.5  They especially focus on the ‘mental processes’ exception. One example given includes a process of ranking things, and notes that there is no reason a computer is required to do the ranking, a human could conceivably do it mentally. Thus, in this case, the invention is deemed ineligible for patent protection.

However, as with the examples granted by the EPO, if the invention provides some kind of output that requires manipulating images, creating a sound file, or anything else that means the process could not be replicated ‘mentally’, then the subject matter is more likely eligible for patenting. An article by law firm Mintz summarises the matter:

  • This may seem an absurd restriction to some, as the human mind might be able to carry out the millions of calculations a neural network can perform, even if there is no guarantee that a human mind could finish those calculations in one lifetime. However, permitting patents on basic calculations would cripple scientific exploration and advancement. Therefore, to be eligible for patent protection, an invention centred on an algorithm must significantly advance a specific technical application, not merely use an algorithm to solve a problem. The patent application must explain in detail how the claimed algorithm interacts with the physical infrastructure of the computer, network, or both and explain the real-world problem the invention is meant to address.6

Even if the answer to this question is ‘yes’ (that is, there is a ‘mental process’ involved) that does not automatically preclude patentability. In these cases, the examiners are being directed to consider if the abstract mental process is ‘Integrated into a Practical Application’ (quoting directly). The way this is phrased in one of the examples is ‘The claim as a whole integrates the mental process into a practical application’.7

In practice, this suggests that arguing “the process may be one that can be done mentally, but it is being done practically because it involves a computer” is not sufficient to clear this consideration. Simply adding ‘a computer’ is insufficient to demonstrate a practical application.

Underscoring the importance of carefully defining the practical application by being descriptive of the problem being solved, IP Watchdog provides this quote:

‘In a nutshell, if you are going to write a patent application in such a way that the reader will be left wondering what the innovation is, what the problem being solved is, or the technical particulars on how the innovation actually solves the problem and achieves the specified functionality, you should not expect a patent’.8








Certica Solutions Inc 



Method and apparatus for performing dynamic textual complexity analysis using machine learning artificial intelligence 


Well Checked Systems International LLC 



System and method for machine learning predictive maintenance through auditory detection on natural gas compressors 


Facebook Inc



Generating personalized content summaries for users 

The first example satisfies the previously discussed criterion of being a very concrete application of machine learning. Although it could be argued that analysing text is a process that could be performed purely ‘mentally’, the claims do describe supplying predictions as an output, which may be enough to meet the conditions of the new USPTO guidelines.

The second example is perhaps a more ‘classic’ case of a defined ‘practical’ problem (in this case, using sensors to predict maintenance) having a solution implementing some form of machine learning. In the claims, some steps are nebulously described as ‘classifying’ auditory signals or the system being able to ‘determine’ anomalies: the implication being that these may be performed by ML algorithms. As with the third example of EPO patents in the previous article, this is not made explicit in the claims themselves.

The third example is perhaps the closest to something that may be regarded as ‘abstract’ as a result of being something that could be achieved through ‘mental’ calculation, being principally concerned with selecting content objects based on a user request, analysing them, and summarising them. However, even in this case the fact that there is a clearly defined output (the summary itself) that can then be presented to an end-user suggests that there is sufficient ‘practical application’ of the invention for the USPTO. The same can often be seen in other AI/ML technology applied to social media or online advertising.

Filing Data

Worldwide, applications for AI and ML patents continue to grow approximately exponentially, with a compound annual growth rate of around 20%. Funding remains strong, with an estimated $16.6bn of VC investment in AI focussed start-ups in 2019 in the US alone, and roughly $11bn in the first three quarters of 2020 despite the pandemic.9  In the graphs below, the shaded area represents incomplete data from recent filings.

‘Big Tech’ companies dominate in AI patents, with Microsoft showing a commanding lead. In recent years, Chinese companies such as Baidu and the Alibaba Group have shown increased filing, in line with increased filing from China as a whole.

Terminology used in the patent texts indicates a focus on ‘classical’ computer science problems such as image processing and sensing as well as a very sharp recent increase in filing around the field of biotechnology. Drug discovery (designing or repurposing drug molecules for new treatments using AI) is a fast-growing area and has received a large boost in funding in recent years, as well as increased visibility due to the Covid-19 pandemic.10

Vehicles, mobile devices, and robots were mentioned most frequently with regards to hardware. This reflects trends in the development of autonomous vehicles and robotics, as well as an increase of AI deployments on mobile phones.

Future Actions

The subject of patenting AI/ML is an active and ongoing issue in most patent offices worldwide. The EPO’s recent conferences highlighted several future challenges, and the USPTO has indicated they may make additional changes or clarifications. As such, it is important to keep abreast of updates or new information from these authorities as they are released. SMEs can expect ongoing interest from ‘big tech’ companies, and should be mindful of their freedom to operate within AI and ML in light of the scale of some of the portfolios of these large corporations especially as they get nearer to a significant funding round, trade sale or IPO.

Prior art searches remain a vital component of qualifying potential patent filings, especially as ML and AI are not always mentioned explicitly in patent claims, but may fall under general umbrella terms such as method steps involving “analysing” or “extracting information” or “classifying”. In addition, landscape reports of the current patent environment can provide invaluable information for two major reasons. Firstly, they can serve to highlight as-yet untapped areas of applications for AI and ML that may be exploited with future filings. Secondly, they can be a valuable source of knowledge of which patents are currently being granted: examining successful patents may help with the construction and scope of future applications. The findings may also serve to highlight important distinctions in different jurisdictions that could prove the difference between a narrow, low value patent and a high value asset that helps protect a company’s core technology.


1 Supreme Court of the United States, “ALICE CORPORATION PTY. LTD. v. CLS BANK INTERNATIONAL ET AL.”, June 19, 2014

2 Justia US Law, "PUREPREDICTIVE, Inc. v. H20.AI, Inc. Filing 31”, August 29 2017

3 United States Patent and Trademark Office, “U.S. Patent and Trademark Office announces revised guidance for determining subject matter eligibility”, January 4 2019

4 United States Patent and Trademark Office, Public Views on Artificial Intelligence and Intellectual Property Policy Executive Summary p.iii. (accessed November 8, 2021)

5 United States Patent and Trademark Office, Subject Matter Eligibility Examples 37 to 42, (accessed November 8, 2021)

6 Terri Shieh-Newton and Marguerite McConihe, “Patenting Considerations for Artificial Intelligence in Biotech and Synthetic Biology – Part 2: Key Issues in Patent Subject Matter Eligibility,” Mintz, January 30, 2020.

7 USPTO, Subject Matter Eligibility Examples 37 to 42, p.2

8 Gene Quinn, “How to Patent Software in a Post Alice Era”, IP Watchdog, accessed November 8, 2021

9 Statista, “Artificial intelligence (AI) funding investment in the United States by quarter from 2016 to 2020”, accessed November 8, 2021

10 Jeremy Kahn, “Money is pouring in to A.I.-assisted drug discovery, while fewer A.I. startups are getting VC backing”, Fortune, accessed November 8, 2021

Did you find this useful?

Thanks for your feedback

If you would like to help improve further, please complete a 3-minute survey