Google has announced the end of its Bard generative artificial intelligence (AI) program, relaunching it under the new name Gemini while also announcing access to a new model dubbed Gemini Ultra 1.1 — the first, it claims, to outperform human experts across a range of fields.
"For years, we’ve been investing deeply in AI as the single best way to improve Search and all of our products," claims Sundar Pichai, chief executive officer for both Google and its parent company Alphabet. "In December, we took a significant step on our journey to make AI more helpful for everyone with the start of the Gemini era, setting a new state of the art across a wide range of text, image, audio, and video benchmarks."
While December may have been the launch of the "Gemini era," for end users little changed: the primary way users interacted with Google's generative AI offerings was through Bard, a large language model (LLM) chatbot that aimed to compete with OpenAI's popular ChatGPT. Now, though, Bard is no more — supplanted by Gemini, under which branding the service will be made available in a dedicated Android app and on the Google app for Apple's iOS.
"The largest [Gemini] model, Ultra 1.0, is the first to outperform human experts on MMLU (massive multitask language understanding), which uses a combination of 57 subjects — including math, physics, history, law, medicine and ethics — to test knowledge and problem-solving abilities," Pichai claims of the company's progress in its battle with OpenAI. "Today we're taking our next step and bringing Ultra to our products and the world."
Gemini Ultra 1.0, Pichai explains, is being made available to end users for the first time as Gemini Advanced — replacing the smaller Gemini Pro 1.0 model, but only for those subscribing to the new Google One AI Premium plan. "With our Ultra 1.0 model," Google's Sissie Hsiao, vice president and general manager for Gemini experiences and Google Assistant, claims, "Gemini Advanced is far more capable at highly complex tasks like coding, logical reasoning, following nuanced instructions and collaborating on creative projects.
"Gemini Advanced not only allows you to have longer, more detailed conversations; it also better understands the context from your previous prompts."
These features, Google claims, will make Gemini Advanced suitable for use as everything from a coding partner to a personal tutor or creative muse — though the company has not detailed how it is addressing the hallucination problem, by which LLMs can prioritize inventing details to form a best-fit answer over providing actual facts.
Gemini Advanced is now available for those subscribing to the Google One AI Premium Plan at $19.99 per month, with a two-month free trial available for new subscribers. Gemini will also be made available in Google Workspace and Google Cloud, Pichai has confirmed, replacing the existing Duet AI platform.