The AI patent litigation war has not started yet. There is still time for companies to prepare.

Sidley Austin LLP's Alexis Cohen published a piece in Bloomberg Law that should concern every legal team representing companies involved in AI. More than 32,000 patents in CPC subclass G06N (neural networks and machine learning) were granted in 2025, up from 426 in 2008. Yet only 53 district court suits asserting patents in that classification were filed last year.

This gap between patent issuances and assertions signals that the expected wave of litigation has not yet arrived. Cohen draws a parallel to smartphones, where litigation surged years after launch. If AI follows a similar pattern, the litigation landscape will look very different within the next few years.

What Companies Can Be Doing Now to Prepare

  • Evaluate trade secret protection for algorithms, training data, and models. In Applied Predictive Technologies v. MarketDial, the Federal Circuit affirmed summary judgment against a plaintiff that could not sufficiently identify its trade secrets. Thus, a trade secret that cannot be adequately defined cannot be protected.
  • Assess portfolio scope. Are there gaps for technology that should be covered? Are ownership and assignment records sound?
  • Evaluate claims involving generic machine learning applications for patent eligibility. In Recentive Analytics v. Fox Corp., the Federal Circuit held that applying generic ML to a new data environment, without disclosing improvements to the machine learning models themselves, does not make a claim eligible.
  • Audit documentation now. Has the development team recorded how problems were solved and how iterations were tested? In litigation, that record is everything.
  • Align IP strategy with marketing strategy. Marketing descriptions that conflict with IP strategy create exposure that can be hard to correct later.

Legal strategists should consider that patent owners have fared better at the PTAB — where challenges are often limited to §§ 102 and 103 prior art — than in district court, where § 101 eligibility can also be raised. This makes eligibility a noteworthy future risk, one that well-prepared specifications and claims today can help mitigate.

For early-stage companies, filing provisionals now — with technically grounded specifications — establishes priority and can serve as proactive risk management before threats escalate.

AI litigation looms. Companies that prepare early will be best positioned when it arrives.