OpenAI security and legal flare‑ups

OpenAI spent the past week answering two very different threats: a software supply-chain scare and an escalating court fight with Elon Musk. (openai.com) On April 10, OpenAI said a third-party developer library called Axios was compromised on March 31 in a broader industry incident, and that it was taking steps to protect the certificates that verify its macOS apps are genuine. The company said it found no evidence that user data, internal systems, intellectual property, or shipped software were compromised. (openai.com) Reuters reported the Axios issue was linked to a wider software supply-chain attack, with OpenAI urging macOS users to update to the latest versions of its apps after rotating security certificates. CNBC separately reported the company said the incident did not expose customer data. (tech.yahoo.com, cnbc.com) A software supply-chain attack works like tampering with a part before it reaches the factory: attackers target a widely used tool so the damage can spread downstream to many companies at once. In OpenAI’s case, the practical risk it flagged was trust in the signing process for its Mac software, not a confirmed breach of ChatGPT accounts or model data. (openai.com, tech.yahoo.com) At the same time, OpenAI is preparing for an April 27 trial in Musk’s lawsuit over the company’s shift from its original nonprofit structure toward a for-profit model tied to Microsoft financing. Musk sued OpenAI and Microsoft in 2024, arguing the company abandoned its founding mission; OpenAI has said his claims misstate the company’s structure and motives. (bloomberg.com, engadget.com) In a filing reported April 11, OpenAI accused Musk of a last-minute “legal ambush,” saying he changed what he was seeking just weeks before trial. Bloomberg reported Musk’s new proposals included sending any trial recovery back to OpenAI, unwinding its conversion, and giving the court oversight over financings and transactions. (bloomberg.com, engadget.com) OpenAI said those proposals were “legally improper and factually unsupported” and would require different evidence and witnesses than the company had prepared. Reuters, in a separate April 6 report, also said OpenAI had asked the California and Delaware attorneys general to investigate what it called Musk’s “improper and anti-competitive behavior.” (engadget.com, msn.com) Musk helped found OpenAI in 2015, left its board in 2018, and has since become one of its loudest critics while building a rival artificial intelligence company, xAI. That history is why a court fight over OpenAI’s structure now sits alongside a security response over how its software is built and signed. (engadget.com, bloomberg.com) For now, OpenAI’s public line is narrow on both fronts: update the Mac apps, and get ready for the April 27 courtroom showdown. (openai.com, engadget.com)