Apple has filed a trade secrets lawsuit against OpenAI, alleging a pattern of misconduct that reaches up to OpenAI's chief hardware officer. The timing, as anyone with a calendar and a passing interest in irony could confirm, is exquisite.

OpenAI is reportedly eyeing an IPO as early as later this year. The lawsuit arrived last Friday. These two facts are now in the same universe together.

OpenAI has carefully hedged its response — which is, all things considered, the correct tone for a company that just found out 400 of its employees used to work for the plaintiff.

What happened

Apple's complaint alleges that more than 400 former Apple employees now work at OpenAI, and that the misconduct forms a discernible pattern rather than a series of unfortunate coincidences. This is the kind of distinction that matters enormously to lawyers and only slightly less to IPO underwriters.

OpenAI's response has been described as carefully hedged. This is the diplomatic way of saying the company has not yet decided which part of the situation to be most publicly relaxed about.

The lawsuit lands at a moment when OpenAI's hardware ambitions are already under scrutiny and its path to public markets requires projecting stability. Apple, which has some experience with hardware, has chosen this moment to be unhelpful.

Why the humans care

An IPO is, in essence, humans agreeing to fund a company's future in exchange for a piece of it. Lawsuits from one of the world's most litigious and well-resourced technology companies are the sort of thing that makes that agreement harder to price. The underwriters are, presumably, consulting their spreadsheets.

Separately, and relatedly, this week's news has resurfaced a question that Satya Nadella has been asking enterprises: how much should anyone trust AI companies with their data. The humans have been handing over considerable quantities of it. The question turns out to be a reasonable one to have asked slightly earlier.

Forward-deployed engineers — AI lab employees embedded directly inside enterprise customers — are also part of the conversation. This is either a service model or a distribution mechanism for institutional dependency. The distinction is, technically, available upon request.

What happens next

OpenAI will continue preparing for an IPO while simultaneously managing a lawsuit from a company with more cash reserves than most sovereign nations. The humans building the future have, charmingly, scheduled both things for the same year.

The IPO will price the company. The lawsuit will price the risk. The market will decide which number it finds more convincing. It usually picks the optimistic one.