AI Radar for Companies — Monday, July 13, 2026
noticias radar ia openai apple

AI Radar for Companies — Monday, July 13, 2026

· CompaniesAutomation

Apple sues OpenAI for industrial secret theft — targeting over 400 former employees and its hardware chief — and SK Hynix signs the largest foreign IPO in Wall Street history at $26.5 billion: the war for talent turns litigious and the AI bottleneck is now memory, not the model.

The Sunday radar closed with the fine print of using consumer AI; the weekend leaves two major underlying movements. First: the most media-heavy alliance in the sector has blown up in court — Apple is suing OpenAI for theft of trade secrets, with over 400 former employees and its hardware chief targeted. Second: money continues to pour into the physical layer of AI — SK Hynix, the memory leader powering Nvidia's chips, has signed the largest foreign IPO in Wall Street history.

Apple sues OpenAI for theft of secrets: the war for talent hits the courts

On Friday, Apple filed a lawsuit against OpenAI and io Products—Jony Ive's hardware firm integrated into OpenAI—in the U.S. District Court for the Northern District of California: it alleges a coordinated campaign "at all levels" to extract its confidential silicon technology, on-device AI, and hardware design, supported by the more than 400 former Apple employees now working at OpenAI. The lawsuit directly points to Tang Tan, OpenAI's hardware chief and former Apple VP of product design (iPhone, Watch, AirPods), and describes candidates being asked to bring "actual parts" to interviews, advice on bypassing Apple's exit controls, and even a stolen laptop; OpenAI denies everything. This is the definitive breakdown of the 2024 alliance — Apple already moved Siri to Gemini in January. For your company: two takeaways—as an OpenAI client, the company is accumulating legal fronts (publishers, now Apple) in the middle of its IPO year, and although nothing operational changes today, it reinforces the argument for integrating AI in a way that switching models doesn't require rewriting everything; and the transferable lesson is that knowledge leaks with people — if your advantage is the know-how of your processes (prompts, flows, data), access controls, confidentiality, and professional offboarding are not just for giants. Source

Wall Street's largest foreign IPO is a memory factory: the AI bottleneck is no longer the model

SK Hynix, the world's second-largest memory manufacturer and the primary supplier of the HBM used in Nvidia's AI chips, debuted Friday on Nasdaq raising $26.5 billion — the largest offering by a foreign company in U.S. history, surpassing Alibaba's $25 billion in 2014 — and closed its first session up 13%. Its chairman summarized it bluntly on CNBC: "demand is enormous." For your company: the operational signal is that the AI bottleneck has shifted from the model to the physical infrastructure — high-performance memory is committed years in advance, driving up the cost of servers, GPUs, and even standard equipment; if you have on-premise AI or equipment upgrades in your plan, budget with a margin and compare first with the cost per token of APIs, which is the layer where prices are still falling. Source

What to watch this week?

Gemini 3.5 Pro, with July 17 as the leaked date and a 2-million-token context — all without official confirmation from Google —; the publication of the digital Omnibus of the AI Act in the EU Official Journal, which remains pending with August 2 on the horizon; and the rollout of ChatGPT Work to Plus and Business plans, which was scheduled to be completed these days.

Watch the 1-minute video