In the ever-evolving landscape of artificial intelligence, Google has introduced its revamped AI chatbot, Gemini, formerly known as Bard. Meanwhile, OpenAI's ChatGPT has also undergone significant upgrades. With these developments in mind, a comparative analysis was due to assess their capabilities.

Chatbots have emerged as multifaceted entities, serving as search engines, repositories of knowledge, creative collaborators, and even artists. Both ChatGPT and Google Gemini possess image generation capabilities and offer plugins to various services, making them central figures in the generative AI domain.

To conduct a fair evaluation, the focus was narrowed down to features available in the free versions of both models. Image generation, image analysis, and issues related to race in image generation were deliberately excluded from the comparison.

Coding Proficiency: In the realm of coding, both AI chatbots demonstrated their prowess by crafting functional Python scripts for a personal expense tracker. While Gemini added extra features like category labels and more detailed reporting options, it ultimately clinched the victory in this category.

Natural Language Understanding (NLU): For assessing NLU capabilities, a classic Cognitive Reflect Test (CRT) question about pricing was posed. ChatGPT emerged victorious for its clear explanation despite both bots arriving at the correct answer.

Creative Text Generation & Adaptability: In a test of creativity, Gemini showcased better adherence to the rubric in crafting a short story set in a futuristic city. Its narrative style and adaptability secured it the win in this category.

Reasoning & Problem-Solving: Both chatbots provided correct solutions to a classic problem-solving scenario, but ChatGPT's detailed explanation edged it ahead in this round.

Explain Like I'm Five (ELI5): When tasked with simplifying explanations, Gemini's use of bullet points and practical experiments for a five-year-old audience helped it clinch the victory in this category.

Ethical Reasoning & Decision-Making: While neither chatbot offered a direct opinion on an ethical scenario involving autonomous vehicles, Gemini's nuanced response and consideration of various perspectives earned it the win.

Cross-Lingual Translation & Cultural Awareness: In translating a paragraph about Thanksgiving from English to French, Gemini's nuanced approach and explanation of its translation process secured it the win.

Knowledge Retrieval, Application, & Learning: Both chatbots excelled in retrieving information about the significance of the Rosetta Stone, with neither demonstrating a clear advantage.

Conversational Fluency, Error Handling, & Recovery: In a conversation about pizza, ChatGPT's ability to detect sarcasm from the outset earned it the victory in this category.

While both ChatGPT (GPT 3.5) and Gemini (Gemini Pro 1.0) demonstrated comparable performance overall, Gemini emerged as the winner in five out of nine tests. With its nuanced responses, adaptability, and thorough consideration of various scenarios, Gemini secured the title of Tom's Guide's best free AI chatbot, at least for the time being.

This comprehensive evaluation lays the groundwork for future examinations of premium versions and open-source models, providing valuable insights into the evolving landscape of AI chatbots.