元エンジニアがバグを悪用して企業秘密を盗んだとして、AppleがOpenAIを提訴Apple sues OpenAI after ex-engineer allegedly used bug to steal trade secrets

Apple has filed a lawsuit against OpenAI, alleging that a former Apple engineer exploited a software vulnerability to steal proprietary trade secrets before carrying that information to the AI company, according to a report from Ars Technica. The case is notable because it pits two of the most prominent names in technology against each other over intellectual property, and it reflects the rising stakes as companies race to build and deploy advanced artificial intelligence.

According to the complaint as described, the dispute centers on a former engineer who is said to have used a bug in Apple's internal systems to access and extract confidential material. Apple appears to argue that this information amounted to protected trade secrets and that it subsequently made its way to OpenAI, where the individual is said to have moved. The framing suggests Apple is treating the alleged conduct not merely as an employee's individual misstep but as a matter that implicates OpenAI as the recipient of the material. The specific technical details of the vulnerability, the nature of the information involved, and the precise claims against each party will likely become clearer as court filings are made public.

Trade secret litigation of this kind typically hinges on whether the information at issue was genuinely secret, whether the company took reasonable steps to protect it, and whether it was acquired or used through improper means. In the United States, such cases are often brought under the federal Defend Trade Secrets Act as well as state law, and they frequently involve former employees who change jobs within the same industry. Companies commonly rely on non-disclosure agreements, confidentiality policies, and access controls to safeguard sensitive work, and allegations that a software bug was used to bypass such controls, if accurate, would be a central element of Apple's argument.

The suit arrives amid intense competition for AI talent, where engineers and researchers routinely move between rival firms carrying deep technical knowledge. That mobility has repeatedly raised legal questions about where an individual's general expertise ends and a former employer's protected information begins. The situation echoes earlier high-profile disputes in adjacent fields, most notably the litigation between Waymo and Uber several years ago over self-driving car technology, which centered on a departing engineer accused of taking confidential files. Those cases illustrate how difficult and closely watched trade secret claims can be, and how they can shape hiring practices across an industry.

Context around the two companies adds another dimension. Apple is well known for a culture of secrecy and for guarding its internal projects tightly, and it has been expanding its own artificial intelligence efforts, including the Apple Intelligence features integrated across its devices. OpenAI, meanwhile, has grown rapidly as a developer of large language models and consumer AI products, and the two firms have had a commercial relationship in areas where Apple has incorporated outside AI capabilities. A legal fight between partners or partial collaborators is not unusual in the technology sector, where companies frequently cooperate in some domains while competing or clashing in others.

It is worth emphasizing that the allegations described remain unproven claims at this stage, and OpenAI has the opportunity to respond and contest them in court. Defendants in trade secret cases often argue that the information was not actually secret, that it was independently developed, or that the departing employee did not improperly use it. How OpenAI characterizes the former engineer's role and what, if anything, the company knew about the alleged conduct will be important to watch.

More broadly, the case underscores a growing tension in the AI industry, where enormous investment, fierce talent competition, and the strategic value of proprietary methods create fertile ground for disputes. As firms pour resources into models, tooling, and infrastructure, the boundaries around what constitutes a protectable secret are being tested more frequently. The outcome, or any settlement, could influence how technology companies structure their security practices, employment agreements, and hiring of engineers from competitors. For now, the filing signals that established players are prepared to use the courts to defend what they consider core competitive advantages.