Samsung Electronics Co. Ltd.’s upcoming new flagship smartphone, the Galaxy S24 series, will be the among the first devices in the world to integrate Google Cloud’s most advanced generative artificial intelligence models, the companies said in an event today.

The Samsung Galaxy S24 smartphone series was announced at Samsung’s Unpacked event in San Jose today, and it will deploy Google’s new Gemini Pro and Imagen 2 models, with processing powered by Google Cloud’s Vertex AI platform in the cloud.

Samsung said the announcement means it will be the first Google Cloud partner to bring Gemini Pro on Vertex AI to consumers. The Gemini model, announced in December, is Google’s most powerful and sophisticated generative AI model, designed to surpass the capabilities of OpenAI’s GPT-4, the engine behind ChatGPT. At the time of its launch, Google explained that Gemini is a natively multimodal large language model, which means it can understand and generate various kinds of information, including text, code, audio, images and videos.

According to Samsung, Google’s Gemini Pro will be integrated within a number of its native applications on the Galaxy S24 smartphones. Users will be able to take advantage of AI-powered summarization and classification features on apps such as Notes, Voice Recorder and Keyboard. Because Gemini Pro is powered by Vertex AI, it will come with enhanced security, privacy, safety and data compliance functionality, the companies promised.

In addition to Gemini Pro, the Galaxy S24 smartphones will also integrate Google’s text-to-image diffusion model Imagen 2, which was built by Google DeepMind. Imagen 2 on Vertex AI will enable smart and intuitive photo editing capabilities within Samsung’s new Generative Edit application. The integration of Google’s most advanced generative AI models underscores Samsung’s belief in making this technology more helpful and accessible, said Janghyun Yoon, a corporate executive vice president at Samsung.

Gemini Pro is a lighter version of the Gemini LLM that’s designed to scale across a broader range of tasks, and is likely powerful enough for the kinds of applications Samsung envisions. However, it’s also working with Google to test Gemini Ultra, the most powerful and capable version of Gemini, which is aimed at extremely complex tasks, though the companies didn’t say what kinds of applications it might be used for.

In addition, the Samsung Galaxy S24 phones will also use Gemini Nano, the most lightweight version of the LLM, to perform certain kinds of tasks locally, meaning processing will be done on the device itself. Gemini Nano is being integrated with the S24 as part of the underlying Android 14 operating system, the companies explained. That means users will be able to access generative AI capabilities even without access to the internet.

The alliance between Samsung and Google is a match made in heaven for both companies, said Andy Thurai, vice president and principal analyst of Constellation Research Inc. For Google, it means it will gain a significant foothold in the mobile generative AI market, while Samsung wins because it can offer more premium features and better compete with Apple Inc., the analyst explained.

“Because Samsung has already chosen Android as its operating system of choice, this alliance and integration is fairly easy,” Thurai said. “A lot of these functions are already available in Google’s Pixel smartphones, so this is not a first in the mobile world. That said, Google’s Pixel phones do not have anything near the same market penetration as Samsung’s devices.”

The alliance is useful for Samsung, the analyst said, because the company cannot compete in the market for lower-end smartphones where numerous affordable handsets from Chinese brands provide strong competition. “Gemini Pro and Imagen 2 on the Samsung Galaxy S24 will help the company to challenge Apple’s iPhone in the premium smartphone segment, where it has recently lost its mojo,” he said.

According to Thurai, the success of this alliance will be closely monitored not only by Apple, but also by Microsoft Corp., which despite being perceived as a leader in the generative AI industry, has a non-existent presence in the mobile area. He believes that the two companies could well explore their own alliance in mobile generative AI, as they are seemingly natural bedfellows.

“They both need each other,” Thurai said. “Although Apple has done some things with AI, such as its facial ID unlock feature that uses facial recognition, its A15 bionic AI chip and Siri’s autocorrect and photo editing features, it doesn’t offer much else. Apple is reportedly spending around $1 billion a year to try and catch up in the AI arms race, and wants to have something to show for it by the time iOS18 is released later this year.”

Google Cloud Chief Executive Thomas Kurian said the company believes there is a tremendous opportunity to use generative AI on smartphones. The technology will power more meaningful experiences that can “stimulate and strengthen connection and communication for millions,” he explained. “With Gemini, Samsung’s developers can leverage Google Cloud’s world-class infrastructure, cutting-edge performance, and flexibility to deliver safe, reliable, and engaging generative AI powered applications on Samsung smartphone devices,” he added.

Image: Google