Ancient Greek sophist Protagoras famously said, \u201cThere are two sides to every question.\u201d But OpenAI\u2019s\u00a0latest attempt at reality distortion seems determined to narrow this dispute to just one. In its motion to reject\u00a0Apple\u2019s complaint,\u00a0the company does not meaningfully acknowledge the criticisms levelled against it, preferring instead to recast the case as a grievance over talent retention and product-market failure.<\/p>\n
In case you missed the news,\u00a0OpenAI filed a motion to the court\u00a0to dismiss Apple\u2019s\u00a0recent lawsuit against it. In that filing, OpenAI argued that, \u201cApple should not be permitted to use a baseless and pretextual lawsuit to make up for its shortcomings in the market for talent and retaining its employees, and its failures to integrate AI into its products.\u201d<\/p>\n
The company\u2019s dismissal claims Apple\u2019s case was, \u201cplainly filed without adequate investigation and built on selectively excerpted communications and ordinary conduct stripped of context,\u201d adding, in a turn of phrase borrowed from Apple\u2019s own complaint, that it is \u201crotten to its core.\u201d<\/p>\n
As ever with litigation, these are allegations and counter-allegations rather than findings of fact. The value of the filings is that they show how each side wants the court, and the public, to understand the same disputed events. At the moment, we don\u2019t yet know how Apple will respond to OpenAI\u2019s response; it follows that company\u2019s failed attempt to woo public opinion earlier in the week when it deployed what some see as a\u00a0\u201ccookie jar\u201d defense, arguing that Apple\u2019s secrets only slipped out because the figurative jar lid was open.<\/p>\n
OpenAI likely hopes for more success with its latest attempt to defend itself against Apple\u2019s claims it engaged in a coordinated attempt to obtain trade secrets through questionable recruitment practices.<\/p>\n
Central to the company\u2019s counter-argument are its attempts to recharacterize some of Apple\u2019s claims. For example, Apple alleges that one former staffer, Chang Liu, downloaded confidential files after leaving the company. OpenAI argues that Liu was instead attempting to help ex-colleagues who asked him for assistance. This is a useful example of the Protagorean frame: both companies are trying to extract different meanings from the same event.<\/p>\n
The courts will need to decide which version of events is closer to the truth. What is already clear is that OpenAI has been actively involved in recruiting Apple staff, including the services of former Chief Design Officer Jony Ive, as it develops a product that, to a layman like me, sounds likely to compete with Apple hardware. OpenAI says those recruitments reflect Apple\u2019s failure to retain its staff; Apple argues its competitor is using exfiltrated confidential information to guide its hiring. The court will need to decide that story as well.<\/p>\n
Ultimately, I don\u2019t expect OpenAI\u2019s efforts to have the court reject Apple\u2019s lawsuit to succeed.\u00a0Apple is asking for discovery\u00a0precisely so it can test whether its reading of this distorted reality is supported by OpenAI\u2019s internal procedures and the available facts. One of OpenAI\u2019s arguments seems to be that Apple has not researched the matter thoroughly enough; Apple is quite literally requesting discovery to do just that.<\/p>\n
I don\u2019t know what discovery might turn up, but it does amuse me to think Apple could build its own large language model to boost the discovery process and identify communication conduits that might otherwise be obscured in the evidence initially available to it. How high, and in what direction, do OpenAI\u2019s claimed recruitment practices go, and who is implicated in them? That\u2019s something we might find out in the coming months.<\/p>\n
OpenAI\u2019s Protagorean defense extends a little further, of course, as the company also said it had \u201cno use, need or desire for Apple\u2019s trade secrets\u201d because it is building \u201csomething entirely new.\u201d This may surprise Apple, which has already alleged that OpenAI contacted its\u00a0manufacturing partners\u00a0and sought access to secret manufacturing processes Apple developed with them.<\/p>\n
Once again, it will be up to the courts to decide whether those events took place, or if OpenAI\u2019s defense has substance. Given that this dispute centers on product design and involves the AI company\u2019s growing army of former Apple design and development staff, I find the denial hard to accept. But courts tend to make their own decisions, for good, or for ill.\u00a0<\/p>\n
What happens next? I think this attempt to reject the original litigation will fail, which means the case will enter the discovery process before one of two outcomes becomes more likely: A bitter public battle that lasts for years and might well end up in the Supreme Court, or an out-of-court settlement shaped by which side gains the most compelling evidence.<\/p>\n
Like any war, there are really only two options: one side fights until the other can no longer continue, or both sides find a way to settle. The path to settlement may begin by recognizing that two stories can be applied to the same facts, and that the version closest to the truth often sits somewhere between them. I\u2019m not a lawyer and I don\u2019t have insider insight into the practicalities of the case, but based on what has been revealed so far, the most plausible combined story may be that Apple\u2019s own vulnerabilities helped create an environment OpenAI chose to exploit. If so, Apple\u2019s legal team will be searching hard for evidence of intent.<\/p>\n