Google's new Gemini AI model can show you that it's...
In September, OpenAI rolled out its o1-preview for ChatGPT, a new series of models designed with strong reasoning capabilities to deliver more thoughtful answers rather than speedy answers. Since then, multiple companies have rolled out their reasoning AI models, such as the DeepSeek-R1 model and Alibaba's QwQ-32B-Preview model.
Introducing Gemini 2.0 Flash Thinking by Google
Google has now introduced its own reasoning AI model called Gemini 2.0 Flash Thinking. According to Jeff Dean, Chief Scientist for Google DeepMind, this experimental model explicitly shows its thoughts and is trained to use thoughts to strengthen its reasoning. This approach allows the AI to recognize and correct mistakes more easily as it forms answers.
TechCrunch tested the Gemini 2.0 Flash Thinking Experimental model and found mixed results. While the model excelled in solving complex queries, it struggled with simpler questions, like determining the number of R's in the word "strawberry."
Seeing Gemini 2.0 Flash Thinking in Action
Google product lead Logan Kilpatrick shared an example of Gemini 2.0 Flash Thinking solving a challenging puzzle involving both visual and textual clues. The AI model's ability to explain its entire thought process from start to finish impressed many viewers.
While the model's detailed reasoning process may lead to slower response times compared to non-reasoning AI models, the depth of understanding it provides can be valuable, especially in fields like coding, physics, and math. Users can gain insights into their own thinking process and potentially identify areas where they may have erred when seeking assistance from the AI.
Although there is still progress to be made in refining reasoning AI models, the steps taken by companies like Google in developing models like Gemini 2.0 Flash Thinking signal positive advancements in the field of AI technology.
