← All articles

Apple Details Methods Alleged in OpenAI Trade Secret Theft

Apple's lawsuit alleges multiple methods used to steal trade secrets: Chang Liu retained a company laptop and exploited a security bug to access files, while Yu-Ting Peng transmitted confidential information, and Tang Yew Tan solicited project details during job interviews. The complaint also alleges OpenAI coached employees on evading security and misled suppliers about proprietary techniques.

AI-assisted reporting

The essentials

Apple alleges OpenAI used laptop retention, authentication exploits, and employee coaching to systematically steal trade secrets.

Apple's lawsuit reveals specific tactics allegedly used to extract confidential information. Chang Liu allegedly retained an Apple computer and exploited an authentication bug to access restricted files before joining OpenAI. Yu-Ting Peng allegedly transmitted Apple trade secrets to Liu while employed at Apple, facilitating the transfer of proprietary data.

Tang Yew Tan allegedly solicited confidential Apple project information during OpenAI job interviews, leveraging the recruitment process as an intelligence-gathering mechanism. The complaint further alleges OpenAI coached departing Apple employees on evading security procedures, suggesting systematic preparation to circumvent information protection measures. Additionally, OpenAI allegedly misled Apple suppliers regarding authorization for a proprietary metal-finishing technique, indicating attempts to gain access to intellectual property through deceptive representation.

Broaden the view

Other ways to read this story.

Alternative framings generated by SigNews. These are not independent sources or attributed quotations.

Operational Timeline

Sequential Steps in Alleged Data Access and Transfer

The allegations follow a pattern: Liu retained physical access via laptop, exploited technical vulnerabilities to access restricted files, Peng transferred data to Liu while still employed, and Tan extracted information during recruiting conversations. This sequence suggests coordinated steps across multiple individuals, with OpenAI allegedly providing guidance on circumventing security measures to departing employees.

Breach Vectors

Multiple Channels Allegedly Exploited for Information Extraction

Apple identifies distinct attack vectors: technical exploitation through authentication bugs, insider transfer by current employees, recruitment-phase intelligence gathering, and deceptive communication with suppliers. This multi-vector approach suggests sophisticated planning rather than opportunistic access, with alleged roles distributed across individuals positioned at different organizational touchpoints.