Google's Gemini AI Falls Short of GPT-4, but Promises Future Potential

Google’s Gemini AI Falls Short of GPT-4, but Promises Future Potential

Google has been under pressure to keep up with its rivals in the AI race. In May, the company started promoting its new AI model, Gemini, boasting that it could outperform industry leader OpenAI’s GPT-4. However, the version of Gemini that was launched recently does not quite meet that goal just yet. Google’s DeepMind CEO, Demis Hassabis, revealed that Gemini outperforms GPT-4 on various benchmarks, but the catch is that the impressive results are for Gemini Ultra, which won’t be available until next year. The version available now, Gemini Pro, can be accessed through Google’s Bard chatbot.

In a technical paper accompanying the Gemini announcement, it was revealed that Gemini Pro can exceed GPT-3.5 on most measures but falls short of beating OpenAI’s GPT-4. The announcement has received mixed responses from online users. Some have praised Gemini for being the first model to surpass GPT-4, even if only by a small margin. However, others have criticized Google for lack of transparency regarding the training data used for Gemini.

In summary, while Gemini’s launch didn’t meet the full expectations set by Google, it shows promise for future advancements in AI technology. The company plans to release the more capable Gemini Ultra next year, which is expected to provide even better performance compared to GPT-4.

Ads
  

Source: The version of Google’s Gemini you can use right now doesn’t beat OpenAI’s GPT-4

Similar Posts