Why doesn't anybody copy Apple?
by Horace Dediu
Apple accuses OpenAI, and former design star Jony Ive’s io Products firm, of stealing hardware trade secrets in blockbuster lawsuit
Apple sued OpenAI on Friday for allegedly stealing its trade secrets, an extraordinary move that pits the $4.6 trillion iPhone maker against the fast-growing AI startup as it prepares to release a new class of hardware products that have been shaped in large part by Apple’s former design boss.
In the lawsuit, filed in the Northern District of California court, Apple accused two former employees now working at OpenAI of systematically stealing confidential data, including information about unreleased hardware products, technical specifications, and details about vendors and contractors in Apple’s supply chain. The complaint also listed OpenAI as a defendant, as well as io Products, a hardware design firm acquired by OpenAI last year that was co-founded by Apple’s former design boss, Jony Ive.
Source: FORTUNE
NB: Jobs said that if you took anybody out of Apple they would be useless. He was right. In my 2013 post "Why Doesn't Anybody Copy Apple" I agued that even though everyone tries to copy Apple's products nobody tries to re-create Apple's operating model. OpenAI seems to be the first to try, though I'd argue they don't have a clue what they are doing.
What Ive and Altman tried to do was brute force replicate Apple by taking as many people and processes from Apple to replicate the operating model. What they could not replicate was the priorities. Resources are easy to steal or replicate. Processes are much more difficult but with a lot of persistence it may be possible. But Priorities cannot be replicated because they include every aspect of the business logic. OpenAI has a miserable business. They cannot graft a device business any better than Google did and arguably have a much shorter runway to even try. It does appear to be a grab the money and run type venture.
Ive's greed has destroyed his legacy. He will try to distance himself but he is forever tainted.
6 comments
rbgayle ·
IMHO, OpenAI's main problem, besides not really understanding what makes Apple great (OpenAI should have stolen the leaders of Apple University ;-), is the type of people Apple tends to attract and reward. At base, I believe Apple succeeds because it selects for autotelic people, those for whom the journey is the joy, not simply how much they are paid. Internally-driven, autotelics will spend huge amounts of time solving problems because that is what excites them. They drop into the flow really easily, concentrating huge amounts of intellectual and physical energy, much like pro athletes. Sure, money is great, and they expect to get paid, but they will run over hot coals if the problems are interesting; not because they are directed to. But piss off an autotelic (like not recognizing their worth or treating them with disrespect) and they will move their efforts elsewhere, no matter how much money gets thrown at them.
I know because I worked for such a company. People would work long hours not only because they were getting paid, but because they were self-directed and used internal drives to solve problems. They did not need a lot of middle management to tell them what needed to be done. We were innovative, resilient, and very efficient in solving problems that much larger companies failed at accomplishing.
Companies like OpenAI or Meta seem to select specifically for the opposite, for exotelic people - those for whom a reward is the main, or only, reason they work. Instead of being internally driven, they are driven by external forces. Money often fundamentally attracts them to work hard. Sometimes it is a title. They often work hard only if given incentives. Many at OpenAI left Apple mainly because they were getting a huge payday to do so.
These types will do what they are told to do, and little else. They have little internal drive to provide more, like an autotelic will. They are much more likely to wait for others to tell them how to innovate.
So I do not expect OpenAI to eventually prevail. Likely because its lawyers expect to get paid a lot while Apple's well-paid lawyers relish the fight.
Sacto Joe ·
GREAT insight, Richard! Thanks for sharing!
Sacto Joe ·
Hi, Horace. "Ive’s greed has destroyed his legacy. He will try to distance himself but he is forever tainted."
I don't think he was greedy for money or power, but for fame. But in its way, it's just as false a god.
It's also why some people choose to be infamous....
FredHS ·
Thanks, rbgayle for the brilliant analysis.
OpenAI's situation is worse. Benedict Evans wrote an article on them which shows, with data, that they're in a race where leadership changes every few months. And top chasers, like Google and Meta have cashflow to fund them, vs OpenAI's massive debt and equity.
In specific applications, Anthropic beats them.
They are NOT in line to be the 'winner take all'. Yet, Altman insists on a $1T valuation for an IPO that raises $60B when he also says he needs $200B.
Only Apple can deliver private personal cross-platform agentic AI. Hence, a device from Ives doesn't matter. It's a rehash two defunct ideas, dumb terminal to big server, 'the iPhone killer'.
deemery ·
I wouldn't say "Only Apple can deliver"... Google could certainly change direction and move that way. Same thing for Microsoft. But for both of those companies, the change in direction to align along Apple's vector would be massive.
I think it's marginally more likely that a well run AI company (which of course, excludes OpenAI) could also move towards the Apple vector.
If there's to be competition, I'm guessing it would come out of China, with the adaption of a strong Chinese (open source?) AI model with Chinese manufacturing. The challenge there would be the corporate integration. I'll also note in passing that a company coming out of China would never (pending a revolution in China) match Apple's privacy commitments. (I'll also admit to being worried about -Apple's own commitment-, particularly with adding adds to Apple Maps. That's the camel's nose under the privacy tent.)
asymco-admin ·
Great points.