Apple’s decision to rely on Google Gemini for Siri’s language and analytics capabilities in the future is the dominant topic in the industry right now. For Google, the step is considered a great success, as it potentially brings the models to hundreds of millions of other users and receives a kind of accolade. For Apple, on the other hand, the priority was to quickly offer a usable solution, even if this was accompanied by the implicit admission that it could not achieve this through its own development. The Financial Times also covered the deal. The publication can therefore roughly confirm previous reports that Apple transfers around a billion dollars a year to Google. OpenAI refused to cooperate
Another statement is also interesting, because OpenAI was also considered a conceivable option. The Financial Times relies on voices within the company, who say that the matter was discussed intensively. However, OpenAI has made the decision not to become a model supplier for Apple. Accordingly, it would not have been a rejection from Apple because the costs were too high, but rather a lack of willingness on the part of OpenAI. Instead, OpenAI came to the conclusion that it would be better to invest the energy in its own projects – in order to be able to stand out from other providers.
It is unclear how intensive the discussions were about this
What remains unknown, however, is whether there were negotiations at an advanced status at all. Although the description at least suggests discussions, there is no evidence of agreements ready to be concluded or concrete contract offers. The timing of the reported rejection is quite interesting, because it probably took place before the proclaimed “Code Red” – when OpenAI CEO Sam Altman ordered to put all his effort into the further development of GPT, but to treat other projects as a secondary priority.
