Unpacking the Core Issues Behind Apple’s Lawsuit Against OpenAI

Today on Decoder, I’m chatting with Hayden Field, a senior AI reporter, about the significant trade secrets lawsuit involving Apple and OpenAI and what it might mean for OpenAI’s future.

Most of you are likely aware of Apple’s allegations in this case. The company claims that several former Apple employees at OpenAI sought out Apple’s trade secrets during job interviews and even downloaded sensitive files from Apple’s servers related to hardware manufacturing. These accusations are quite serious, though OpenAI has denied them, and we’re still waiting for their legal response.

Hayden has been following OpenAI for years and recently spoke with several lawyers specializing in intellectual property and trade secret law to gauge the seriousness of this case. The outlook isn’t entirely favorable for OpenAI, especially considering its $6.5 billion acquisition of Jony Ive’s AI hardware startup, io Products, back in 2025.

As Hayden explains, Apple is known for being a fierce litigant, and there’s plenty of tech industry history to support that reputation. However, previous cases were against giants like Microsoft and Samsung, companies that could handle the costs and distractions of such litigation. OpenAI, on the other hand, is still in a cash-burning phase and is somewhat directionless as it competes in both enterprise and consumer markets, frequently changing its executive lineup. We spent a lot of time discussing what this lawsuit could mean for a company that has been pivotal in the early AI boom but seems to be facing ongoing crises.

Now, let’s dive into the conversation with Hayden Field about the Apple-OpenAI lawsuit and its potential impact on the AI industry.

This interview has been lightly edited for length and clarity.

Hayden Field, welcome back to Decoder.

Thanks! Great to be here.

Once again, we’re discussing OpenAI.

It never stops.

Today, we’re focusing on whether OpenAI can maintain its focus amid significant distractions, which seems to be a recurring theme for the company. This time, however, the distraction stems from external sources, possibly due to some of their own actions.

We’re talking about the Apple trade secret lawsuit, which feels like a major issue. Apple is accusing OpenAI of stealing various types of information, even during job interviews, where candidates were allegedly asked about their knowledge of Apple products. Can you give us a brief overview of this lawsuit and what’s happening?

Absolutely. It’s quite a situation. I wrote a piece last week titled “Sam Altman didn’t need another lawsuit,” and that sums it up. The claims Apple is making are quite serious. A former Apple employee allegedly solicited trade secrets during interviews, asking candidates to bring hardware outside the Apple office for a “show and tell,” among other things.

The allegations are wild, but when I spoke with experts, they mentioned that while these claims are extreme, they’ve seen similar situations before, just not all in one case involving two major players.

Some of these claims are indeed shocking if proven true. However, experts in the field noted that trade secret cases often involve such tactics; what’s unusual is the scale and the involvement of two large companies.

It’s interesting to note that in the ‘90s, Apple pushed copyright law to its limits by suing Microsoft over the look and feel of Windows. Despite that, Windows is still around, and Microsoft is doing just fine. In the 2000s, Apple took on Samsung over patent violations related to Android and Galaxy phones, resulting in Samsung paying about $1 billion in damages, but Android remains a strong competitor.

Now, we see Apple moving into trade secrets with OpenAI. My main question is whether OpenAI can afford to pay up like Microsoft and Samsung did. Samsung’s $1 billion penalty to launch Galaxy phones was a worthwhile investment. But OpenAI may not be in the same position, especially if Apple’s claims about its former employees pushing boundaries are accurate.

The claims in the lawsuit are quite outrageous, including messages where one individual allegedly accessed confidential Apple information and remarked, “LOL! So funny that I can access this.” There’s a lot to unpack.

I’ve been pondering the same thing. It really hinges on Apple’s willingness to be either petty or logical. The lawyers I spoke with emphasized that Apple is known for being a relentless litigant, as evidenced by past cases. They have the financial resources to pursue this aggressively, so it depends on how angry they are.

They might settle if OpenAI offers enough, but can OpenAI afford that? They’re already having a challenging year, preparing for an IPO, with investors scrutinizing their financials and profitability. It’s a tough time to be shelling out a significant amount to Apple. We’ll have to see how this unfolds.

Many lawyers I spoke with believe this will go to trial and could drag on for years. However, there’s a possibility they could find a way to gather the necessary funds to settle, and Apple might accept it.

The Samsung patent cases dragged on for ages. I remember asking Tim Cook at a Code Conference about the purpose of those lawsuits, and he claimed they were addressing licensing fees for standard essential patents. But it was clear that the real motivation was frustration and a desire to win the lawsuit, regardless of the nominal reasons.

That’s a crucial aspect here. OpenAI has hired several former Apple employees, including Jony Ive, the renowned designer, who is now collaborating with OpenAI on hardware. Tang Tan, OpenAI’s chief hardware officer, has a reputation for being somewhat reckless, according to industry insiders.

He seems to be a central figure in this narrative, with the allegations suggesting that he and others may have crossed lines that feel personal, rather than just corporate. Can you elaborate on Tang Tan and the actions of these former Apple employees that seem to go beyond typical corporate disputes?

Regarding your conversation with Tim Cook, I genuinely believe people underestimate the pettiness of companies. We’ve seen it in recent years, like with the OpenAI-Elon Musk lawsuit. Many issues could have been avoided if companies weren’t acting out of spite, which is understandable given their resources.

As for Tang Tan, he spent 24 years at Apple, serving as the VP of the Apple Watch before leaving in 2024 to join Jony Ive’s hardware company, io Products, which OpenAI later acquired. OpenAI appointed him as chief hardware officer, giving him a significant role after his long tenure at Apple. According to Apple’s lawsuit, he allegedly orchestrated much of the trade secrets gathering during interviews, asking candidates about confidential projects, which may have led them to believe they could share information freely.

There were also instances where he asked candidates to bring hardware components outside the Apple office for a “show and tell.” It’s intriguing to see these claims, as it suggests he may not have been concerned about the consequences if they turn out to be true.

Another key figure is Jony Ive, Apple’s iconic former designer who worked closely with Steve Jobs. He reportedly has a strained relationship with Apple, which he has made clear in various interviews. Now, he’s collaborating with Sam Altman on new devices, even appearing in a video together that felt almost like a romantic coffee date.

Absolutely.

In that video, they referred to phones and laptops as legacy devices.

It quickly became a meme sensation.

It was hilarious. They even shut down half of San Francisco to film it. Ive’s comment about phones and laptops being legacy devices felt like a direct jab at Apple, considering he was instrumental in creating those products.

While we can debate the validity of that statement, it’s noteworthy that Jony Ive isn’t mentioned in the lawsuit. Apple seems to have drawn a line there, avoiding any mention of the fact that it was Ive who went to OpenAI and began hiring his former colleagues, selling a significant portion of his company to OpenAI, and now they’re allegedly involved in stealing trade secrets. Why do you think that is?

Is it because he’s too famous? Too powerful? Or perhaps because he’s still connected to Laurene Powell Jobs? What’s your take?

I suspect it’s more about him knowing how to cover his tracks. It could be that he’s too influential, but I believe he has others doing the heavy lifting for him. The other allegations in the lawsuit are quite serious, and if they had strong evidence against Ive, I think they would have included it. There’s clearly some tension between them, so if they had something substantial, it would likely be in the lawsuit.

Or maybe he’s preoccupied with designing unpopular Ferraris or whatever else he’s up to, but he’s not implicated here. It’s significant because the reason Sam Altman would hire Jony Ive is to lend credibility to whatever new device they’re developing. If they’re aiming to create the next big AI-powered device, having the most renowned designer on board is a strategic move.

It seems OpenAI also sought Apple’s expertise to develop that device. What do you make of the idea that OpenAI needed Apple’s hardware knowledge, as Apple alleges?

To some extent, whether it’s about trade secrets or simply hiring experienced employees, they clearly needed that expertise. OpenAI has witnessed numerous AI hardware failures and is likely trying to avoid becoming the next cautionary tale. Launching an AI hardware device today can be incredibly risky, and if they want to succeed, they need to recruit top talent.

Hardware development is notoriously challenging, so it makes sense for them to seek out experts with a proven track record. OpenAI has shown, like many tech companies, that they can sometimes be out of touch with consumer desires. They might roll out a product that doesn’t resonate with users. While software allows for easier pivots, hardware requires getting it right from the start.

I’m glad you mentioned consumers because that leads to the core issue for OpenAI. The company aimed to be the go-to consumer AI platform, while competitors like Anthropic have focused on enterprise solutions, which have proven lucrative. OpenAI’s ambition was to challenge Google, possibly through Bing or its own app, and to take on the iPhone.

However, it seems consumers aren’t particularly interested in these products or AI at this scale. With college students expressing disdain for AI at graduations, does OpenAI need to shift its focus to enterprise solutions? Should it invest time in consumer hardware before solidifying its business model?

OpenAI has undergone numerous internal pivots. Its latest shift was to concentrate on key revenue streams, particularly enterprise and coding, which they’ve reiterated multiple times. They’ve discontinued various projects, including Sora and Atlas, which feels like a long time ago, but it was just last year.

Everyone thought they were focusing on enterprise, especially with an IPO on the horizon. However, they had already planned the hardware device before this pivot, and now they can’t afford to abandon it since they’ve invested heavily. They’re committed to making it successful.

Having been in the AI field for six years, I can say that while it feels like a long time, I still don’t see a market for OpenAI’s hardware devices. If I’m wrong, we can revisit this in July 2026. A smart speaker might find some buyers, but I think it will appeal to a niche audience rather than the general public.

I suspect it will be one of those products that a select few will love, while the broader public may not adopt it. Adoption often hinges on social factors; if your friends don’t have it, you might feel left out. There’s a social dynamic at play, especially with smartphones. If you don’t have an iPhone, you might miss out on group chats, which can be frustrating.

When we founded our platform during the smartphone boom, it was clear that people would adopt smartphones. The only question was which company would come out on top. Over time, it became evident that Apple would dominate. We covered numerous so-called iPhone killers, but they all faded away, leaving us with iOS and Android as the main players.

Now, we’re in a similar situation with OpenAI’s hardware device. It will likely run its own operating system, and there are discussions about competitors like Elon Musk’s Starlink and xAI potentially launching their own devices. This raises questions reminiscent of those we asked about new smartphone operating systems years ago. Will it have essential apps like Gmail or Instagram? Without those, it won’t replace existing devices.

The challenge is that the smartphone replaced the dumb phone, so any new device must offer something significantly better. Will OpenAI deliver on that? It seems unlikely, but we’ll see.

Another concern is the number of devices OpenAI is reportedly developing—five hardware devices, with the screenless smart speaker being the first. How can they manage five? Shouldn’t they focus their resources on one? Creating even one successful device is already a tall order, so five seems overly ambitious.

There’s likely a lot in the pipeline, and some may come and go. Jony Ive’s approach to hardware often involves innovating new input methods. The iPod had the click wheel, the iPhone introduced multitouch, and the Apple Watch featured a scroll wheel.

For AI, the ideal input method would be natural language. You should be able to communicate with the device as you would with a person. However, despite several years into the AI revolution, it still doesn’t function seamlessly. This is likely why many AI devices have struggled, as they often fail to deliver on their promises.

Where do you think we stand in terms of understanding AI’s capabilities, and how does that impact hardware development?

We’re still navigating this landscape. Personally, I use AI frequently to test its capabilities, and I’ve noticed improvements over time, though my experiences are anecdotal. Recently, I tried to gather addresses for mapping, and about 60 percent of the results were incorrect, despite the AI assuring me they were accurate.

It’s frustrating when the AI claims to have the right information but fails to deliver. For a hardware device, there are low-stakes questions it can handle, but for more critical inquiries, it’s often easier to look things up yourself rather than risk relying on potentially incorrect answers.

Will this end up like Google Home, where it’s primarily used as a speaker for simple queries? It can only be truly useful if it can be trusted with more significant questions. If it gets something wrong, like flight information, it could lead to real problems.

This seems to be a fundamental issue. The industry claims to have made progress in addressing hallucinations, but whether that aligns with users’ experiences is debatable. Additionally, will OpenAI have access to your data to make the device useful? That raises questions about privacy and competition. Are users comfortable sharing their data with various AI providers? I’m not sure OpenAI is the one people will trust.

Another concern is the growing anxiety around ads. Companies like Anthropic have capitalized on this sentiment with anti-AI ads. OpenAI has faced backlash, trying to assure users that their answers won’t be influenced by ads. However, encountering ads while using ChatGPT can feel intrusive and diminish the sense of privacy.

Trust in AI labs is at an all-time low, and we’re witnessing a rise in AI populism, with backlash against AI leaders. It’s a strange time to launch a consumer-facing AI hardware device.

This lawsuit ties into a broader issue within the AI industry, which often operates on unauthorized data usage. While some may label it theft, the legal landscape is still evolving. The question of whether training data constitutes fair use remains unresolved, and numerous lawsuits could threaten the industry’s foundation.

AI companies have built their models on vast amounts of data, often without permission. For instance, a recent hack revealed that Suno had scraped music from YouTube without authorization, which likely upset both YouTube and the artists involved. Now, Apple is claiming that OpenAI has taken its hardware designs and manufacturing knowledge without consent.

Culturally, this seems to be a recurring theme in the AI industry, where companies believe they can take whatever they need to achieve superintelligence. However, public sentiment is shifting, and artists, writers, and musicians are voicing their concerns, as is Apple.

This lawsuit may serve as a proxy for the larger question of whether companies can take information without permission. Does OpenAI have the cultural capital to defend itself? What can they say?

It’s ironic that the entire AI industry relies on the data it consumes, whether legally or not. AI labs and tech companies often become litigious when others use their information. This irony is particularly pronounced in the AI sector.

The AI industry is known for its close-knit nature, with individuals frequently moving between companies. While it’s common in many industries, the AI sector seems especially interconnected. So, while it’s egregious for former Apple employees to take trade secrets, it’s also worth noting that many companies have thrived on similar practices.

All AI companies get upset when, for example, Chinese firms distill their models. I can imagine a jury hearing Apple’s case and thinking, “You stole music and manufacturing designs, didn’t you? And you get upset when others distill ChatGPT?”

In your reporting, have you encountered any acknowledgment from the AI industry that their frustration over distillation mirrors the feelings of artists and musicians regarding their work being used in models?

I haven’t seen any acknowledgment of that, which is surprising. There seems to be a disconnect. They express outrage over distillation, claiming it violates terms of service, but they don’t seem to recognize the parallel with artists’ concerns about their work being used without permission.

Some listeners might be frustrated with us for not suggesting that Apple is filing this lawsuit out of desperation for falling behind in AI. Apple has struggled to catch up, and now it’s launching its own AI initiatives. Do you think the broader AI industry views this lawsuit as Apple lashing out?

That could certainly be part of it. Apple has been notably behind in AI, and that’s been a source of embarrassment for the company. However, this lawsuit isn’t just about someone stealing trade secrets for Siri; it’s focused on hardware, an area where Apple excels.

While it may benefit Apple to hinder a competitor, the lawsuit’s focus on hardware suggests it’s not merely about AI capabilities. Apple is protecting its strengths in hardware design and manufacturing.

Apple can afford to engage in this litigation, as you’ve pointed out. They can hire a team of lawyers and pursue this aggressively without it significantly impacting their operations. OpenAI, however, is in a different position. They may not have the resources to absorb such a distraction, especially with ongoing executive turnover.

As you’ve been following this story, has it occurred to OpenAI that they need more personnel to handle these challenges? They seem to be losing talent and constantly pivoting, trying to compete with Anthropic, but they can’t afford distractions right now.

They can’t afford it. This is a critical moment in the ongoing AI arms race. With the IPO year upon us, investors are scrutinizing these companies more than ever, looking for profitability or a clear path to it. OpenAI cannot afford distractions at this juncture.

They’re undergoing significant restructuring and executive changes. When I’ve asked them about this, they claim it’s all part of a larger plan, but it doesn’t always come across that way to the public.

Greg Brockman is taking on more control of product development, especially after the departure of Fidji Simo, the AGI chief, due to health issues. However, there have been numerous defections even among regular staff. They need to hire more talent, which requires additional funding, and now isn’t the time for distractions.

There are reports suggesting that OpenAI may delay its IPO. As we speak, SpaceX’s stock price has dipped below its IPO level, indicating that mega IPOs are facing challenges. It might not be wise to proceed without solid revenue or profits. Do you think this lawsuit could push OpenAI to postpone its IPO?

Absolutely. Reports indicate that Sam Altman was initially concerned about filing its Form S-1 before Anthropic. After Anthropic went first, it seems OpenAI reconsidered its timing. Now, with the lawsuit, they might be thinking about delaying even further, especially since they’re unlikely to beat Anthropic to the IPO.

If they can launch a compelling hardware device and make a splash in the consumer market, that would be a strong narrative for the IPO. But if litigation looms over everything they do, it may be best to set those plans aside for now.

That brings me to my final question. My perspective has been that while I understand how AI will function in software and enterprise, the consumer side remains shaky. The large number of free ChatGPT users doesn’t necessarily indicate a strong market for AI products.

People seem to be resistant to AI, as evidenced by negative reactions at graduations and town hall meetings. Can OpenAI bridge that gap? If not, should it focus solely on competing with Anthropic in the enterprise space? What’s your sense from your reporting?

I think they’re aware of the gap to some extent, but not fully. They acknowledge it in speeches and research papers, but I don’t think they grasp how deep the divide is. The emotional aspect complicates things; it’s not just a logical gap. If it were purely logical, they might have a chance to close it, but emotions play a significant role, and I don’t see them bridging that divide.

When I started as an AI reporter six years ago, few people cared about the topic. Recently, however, it’s become a common conversation starter, and people are expressing strong emotions about it. I don’t believe OpenAI will be the company to close that gap.

We’ll see how the hardware situation unfolds. Many creators on platforms like TikTok are trying to shift the narrative, urging people to learn how to use AI tools. While there’s some validity to that, the framing often suggests a steep learning curve, which isn’t always the case.

It’s not that complicated, depending on how you use it. For instance, using Codex for complex coding tasks can be challenging, but chatting with a chatbot is relatively straightforward.

I’d like to pitch you a video series called “Hayden in the Club,” where you engage with people about AI.

I’d love to do that officially instead of just having those conversations informally.

It’s been a while since I’ve been in a club, and it’s certainly changed since my younger days.

Hayden, this has been a fantastic discussion. I have a feeling we’ll be revisiting this lawsuit and OpenAI throughout the year. I look forward to having you back soon.

Can’t wait! Thanks.