When the Supreme Court decided Bilski, we lamented that the “Court regrettably failed to provide guidance in the future about business method patents.” Now we are faced with the result of that failure: a string of cases that leaves us scratching our heads and wondering what, if anything, Bilski meant.

A little background: abstract ideas are not eligible for patent protection under § 101 of the Patent Act. This is fundamental to the patent bargain; without this limitation, parties could obtain harmful monopolies on simple ideas, e.g., ways of running a business or cooking a meal, and be able to (in theory at least) prohibit people from implementing or even thinking those same ideas. In Bilski, the Supreme Court held that the patent at issue, which covered hedging risks for commodity trading, unpatentable because it was too abstract. But the Court also held that other business method patents might be sufficiently concrete. In other words, no blanket exclusions – each should be considered on its own merits.

Unfortunately, this case-by-case approach has made patent litigation even more unpredictable. Just last week, the Federal Circuit issued a ruling in Ultramercial, LLC v. Hulu, a case involving a patent covering a “Method and System for Payment of Intellectual Property Royalties by Interposed Sponsor on Behalf of Consumer over a Telecommunications Network.” More specifically (and in English), the patent claims an 11-step process for distributing “media products that are covered by intellectual property rights protection” that essentially consists of pairing the content with an ad and restricting access until the user views the ad.

The Federal Circuit, faced with the question of whether the patent’s subject matter was too abstract, found that it was not. The Court reasoned that “many” of the “steps are likely to require intricate and complex computer programming,” and, even more troubling, stated that:

In addition, certain of these steps clearly require specific application to the Internet and a cyber-market environment. One clear example is the third step, “providing said media products for sale on an Internet website.” … Viewing the subject matter as a whole, the invention involves an extensive computer interface.

On its face, this all sounds well and good, until one considers the patent’s own depiction of its allegedly not abstract invention:

To be clear, the Federal Circuit only decided the question of whether the patent was impermissibly abstract, not whether it was obvious or not novel or otherwise unpatentable (those questions may still be answered by the lower court).

Ultramercial is not the only patent-eligibility case the Federal Circuit has recently considered. For example, just last month, in Classen Immunotherapies v. Biogen IDEC, the Court found patents for methods of evaluating and improving safety of immunization schedules not impermissibly abstract based on the fact that the patents claimed the actual act of administering the immunization after devising a new schedule.

And just two weeks before its Classen decision, the Federal Circuit issued a ruling in CyberSource Corp. v. Retail Decisions, Inc., in which it went the other way. The Court held that a patent that covered obtaining and comparing intangible data (such as IP and email addresses) pertinent to business risks to detect fraud in credit card transactions over the Internet was ineligible for patent protection because it was too abstract. Central to the Court’s decision was that the process of comparing relevant data could occur with or without a computer, and wholly in someone’s mind – in other words, it was an unpatentable mental process.

Taken together, these post-Bilski cases confuse, rather than clarify, the standard for impermissible abstraction. In four cases (Bilski, Ultramercial, Classen, and CyberSource), two patents were too abstract (patents for hedging risks and detecting credit card fraud) and two were not (patents for showing ads before copyrighted content and devising immunization schedules). For laypeople and attorneys alike, it is hard to understand why the latter two patents were any more concrete than the former. One might argue that the upheld patents required added complexity (computer programming and administering an immunization), but the abstract patents would likewise require additional steps to execute. What distinguishes those steps that are too abstract from those that are not?

Thus it appears patent holders – and the targets of legal threats based on those patents – are in a worse position than ever, unable to make sound predictions about which business method patents are too abstract to be enforced. It is bad enough that innovators must work around patents covering processes that border on the abstract, but it is even worse that those in the field face the uncertainty that has followed Bilski.