In an unexpected development, it appears that Apple is in “active negotiations” with Google to bring its Gemini generative AI model to the iPhone, according to Bloomberg, and the Cupertino company is also reportedly considering using OpenAI’s ChatGPT.

Gemini is Google’s suite of generative artificial intelligence tools, ranging from chatbots to coding assistants. The report’s veracity will be tested when Apple’s next iPhone update, iOS 18, is announced, which is expected to happen at this year’s Worldwide Developers Conference, which usually takes place in June. In fact, it will be the platform to understand what CEO Tim Cook meant by Apple spending “a tremendous amount of time and effort” on AI features.

The Bloomberg report also mentions that Apple recently held discussions with OpenAI and has considered using its model. But a Google deal appears more plausible as the iPhone maker already has a longstanding deal with the Sundar Pichai-led company to keep Google Search the default on its devices. Meanwhile, Samsung has already added a bunch of Galaxy AI-branded features to its Galaxy S24 smartphones. If either company gets to have its AI tools on the iPhone, it means access to over two billion devices.

“What matters here is that Apple may receive or retain some payment from Google for the privilege, especially if it helps Google keep more search revenue,” analysts at Melius Research said in an investors’ note. “For Google, it would be a reputational win versus Microsoft and OpenAI, after a series of missteps in launching reliable AI products.”

Apple may not go with either company as it is apparently testing an in-house chatbot called Apple GPT, and the company reportedly has its own large language model codenamed Ajax.

Nvidia makes progress

Chips aren’t the most interesting tech component to read about but Nvidia, which is now the “king” of the AI world with data centre GPUs and super chips, has revealed Blackwell B200 GPU, the ‘world’s most powerful chip’ for AI, and GB200 “superchip”. Nvidia says the new B200 GPU offers up to 20 petaflops of FP4 horsepower from its 208 billion transistors. Meanwhile, a GB200 that combines two of those GPUs with a single Grace CPU can offer 30 times the performance for LLM inference workloads.

A new dimension to video

Stability AI, a leader in generative AI models, is growing its portfolio with the release of Stable Video 3D (SV3D). The company has been developing video capabilities with its Stable Video technology that enables users to generate short video from an image or text prompt. SV3D builds upon Stability AI’s previous Stable Video Diffusion model. With SV3D one can create and transform multi-view 3D meshes from a single input image. It is available for commercial use with a Stability AI Professional membership and for non-commercial purposes, users can download the model weights from Hugging Face.