Google Gemini AI launched, company calls it smarter than humans in some language work

Google Gemini AI launched, company calls it smarter than humans in some language work

After several months of anticipation and some delay, Google has finally released Gemini, their next-generation artificial intelligence system. Although Google already has Bard, it is anticipated that Gemini AI will be the AI system used to rival — if not surpass — ChatGPT by OpenAI. Google CEO Sundar Pichai described the launch of Gemini as the “biggest science and engineering efforts we’ve undertaken as a company.”

Additionally, Pichai stated that Google’s eight years of AI research has culminated in Gemini. Gemini AI, according to him, will be offered in three modes: Ultra, Pro, and Nano. As the name implies, Gemini will use the largest LLM (large language model) to carry out its AI functions in Ultra. The Nano will utilize the smallest LLM, whereas the Pro will use a smaller LLM. This also raises the prospect that Nano will probably be able to be used locally on PCs and smartphones.

Google wants to compete with other AI systems, especially OpenAI’s ChatGPT, with Gemini. Even though Google led the way in AI through its investment in DeepMind, the company has lagged behind when it comes to actual products. OpenAI, which revolutionized AI last year with ChatGPT, is currently setting the pace in AI. Google may be better equipped to address the issues raised by emerging AI companies thanks to Gemini.

Every change in technology presents a chance to further human progress, enhance scientific understanding, and enhance lives. In a blog post, Sundar Pichai stated, “I think the shift we are seeing with AI right now will be the most profound in our lifetimes, far bigger than the shift to mobile or to the web before it. With Gemini, our most powerful and versatile model to date, we’re expanding on our journey by achieving cutting-edge results on numerous top benchmarks. We’ve made some of the largest scientific and engineering efforts as a company with these new models.”

Demis Hassabis, the CEO and co-founder of Google DeepMind, claimed in the same blog post that Gemini even beats people at a number of language tasks. “With a score of 90.0 percent, Gemini Ultra is the first model to outperform human experts on massive multitask language understanding (MMLU),” the author stated. 

Beginning on Wednesday, Gemini will be integrated into a number of Google services and products, such as Bard, the user-accessible AI system. “Bard will use a fine-tuned version of Gemini Pro for more advanced reasoning, planning, understanding, and more,” according to the company, as of right now.

Leave a Comment

Your email address will not be published.