EPO examiners aren't shy to admit this decline in patent quality and abrupt departure from the European Patent Convention (EPC) [1, 2]

Summary: Aided by Microsoft lobbying (with its very many patent trolls) as well as corrupt Battistelli, the push for software patenting under the guise of "artificial intelligence" ("AI") carries on, boosted by Battistelli's own "Pravda" (which he writes for), IAM Magazine

THE EPO and USPTO both use the buzzword (acronym) "AI" to go about granting software patents. Law firms and their clients are patenting software by consciously invoking this vague buzzword while EPO management sets up events to promote the practice (they also say "ICT", "CII", "4IR" and so forth).

This is not normal. It's not acceptable. They are bypassing laws and misleading everyone. The latest example of software patents being disguised as "AI" to fool patent examiners is about a day old (a press release that says "FICO Awarded Seven New Patents for Analytic Innovations in Fraud Detection and AI," then "FICO has been awarded seven new patents for fraud detection, AI and analytics"). FICO basically gets a bogus -- i.e. likely invalid -- monopoly on abstract ideas (courts would reject these patents), but unless PTAB receives a petition (IPR) FICO can engage in patent shakedown against rivals.

"FICO basically gets a bogus — i.e. likely invalid — monopoly on abstract ideas (courts would reject these patents), but unless PTAB received a petition (IPR) FICO can engage in patent shakedown against rivals."This week's Microsoft- and patent trolls-funded event of IAM also promotes software patents and here's Microsoft doing its 'share' of the task, dubbing software patents "AI".

"Microsoft’s Nicolas Schifano very clear on why much of AI is patentable," IAM wrote (and there's a photo there as well). There's still not a word from them about Battistelli's talk in support of software patents (using the latest buzzword); he was definitely there as scheduled. What's noteworthy is that IAM is now legitimising the judge-bashing Watchtroll. They're all in this together. A lobby of shame.

Unfortunately, words (or terms) like "AI" can be problematic even if examiners recognise these and throw away the papers (applications). Line managers, for example, can use Battistelli's buzzwords du jour (e.g. "ICT", "CII", "4IR") and then demand from patent clerks/examiners that they issue a patent/grant. The way things are going right now are profoundly depressing; the examiners are losing any pride associated with their job (work that is hard to replace because of exit conditions and the nature of contract they signed).

Yesterday we saw AA Thornton & Co's Leonita Paulraj writing about the EPO's patent assessment criteria, which are mostly overlooked/ignored by corrupt Battistelli (who only wants lots of grants in order to game the system). "In the recent past," Paulraj wrote, "the EPO appears to have become more strict with the requirement of plausibility, and therefore it is advisable not to be in a hurry to file an application, especially in the fields of pharmaceuticals and biotechnology, without any clear evidence which supports that the invention has some technical advantages."

She emphasises that it's about pharmaceuticals and biotechnology way upfront:

This article considers how the EPO views the requirement of “plausibility”, which can arise during both the assessment of “sufficient disclosure” of the invention and “inventive step” in proceedings before the EPO, particularly in relation to inventions in the Pharmaceuticals and Life Sciences field.

The EPO Guidelines for Examination (EPO Part F-III, 12) state that if the claimed invention lacks reproducibility, this may become relevant under the requirements of sufficiency of disclosure or inventive step. If an invention lacks reproducibility because its desired technical effect as expressed in the claim is not achieved, this results in a lack of sufficient disclosure, which has to be objected to under Art. 83 EPC (sufficiency). Otherwise, i.e. if the technical effect is not expressed in the claim but is part of the problem to be solved, there is a problem of inventive step, which has to be objected under Art.56 EPC (inventive step).

[...]

The EPO has issued a number of decisions on plausibility, however, none of these decisions are conclusive. The question of plausibility is assessed by the EPO case by case.

From the decisions discussed above, it appears that even if a claim is non-obvious with respect to the prior art or in the absence of any prior art, the EPO could still refuse a case if it is not plausible at the date of filing that the invention does solve the problem it purports to solve.

Likewise, the requirement of sufficiency of disclosure must be met at the date of filing and if the application or patent is found later that it does not meet the requirement, it is no longer possible to correct this deficiency.

In the recent past, the EPO appears to have become more strict with the requirement of plausibility, and therefore it is advisable not to be in a hurry to file an application, especially in the fields of pharmaceuticals and biotechnology, without any clear evidence which supports that the invention has some technical advantages.

From what we can gather, based on what sources tell us, the EPC is being violated routinely in the domain of software, with people like Grant Philpott on top of this colossal failure. Software patents are nowadays being granted routinely by the EPO, often in defiance of examiners' desires and sometimes in part due to buzzwords such as "AI", "ICT", "CII", "4IR". This is going to cause tremendous agony to Europe's already-weak (compared to the US) software industry. The US is meanwhile removing many of these artificial barriers (Alice, which we shall cover in our next post, axes many software patents), spurring a boom in investment for software businesses. ⬆