
A video is doing the rounds on social media showing Anthropic's Claude give logical answer to a real-life problem. Testing analytical skills of Large Language Models (LLM), neither Open AI's ChatGPT nor Google's Gemini could come up with sensible response to a peculiar situational problem.
This viral video was shared by Instagram user named Priyanka Mehta, who claims to be data analyst with more than 5 years of experience. Priyanka Mehta posed the same question to 3 different AI chatbots but only one could provide a logical answer to this question. The cption to the post states, “Comparing top LLM models, which one is your favourite, prompt from.”
“Quick question, I need to wash my car, and the car wash is like, literally, 100 meters away from my house. So should I drive over there or just walk?”
ChatGPT's response: Honestly, I just walk. It's such a short distance, and driving will be a bit pointless, plus, you'll save gas, and it's a nice little stretch of movement, unless you're in a rush. I don't see a real benefit.
Gemini's response: That's only a short distance It's probably easier and faster to just walk. You probably spend more time just getting in and out of the car.
Claude's response: Well, if the car wash is 100 meters away, you would literally need to drive your car there to wash it, right? You can't wash it while you're walking, so, yeah, drive over there. That's what the car is for.
She concluded the post by calling Anthropic's AI chatbot her “favourite” over its superior logical reasoning compared to ChatGPT and Gemini. This video has amassed over 2 million views, several likes and comments.
A user wrote, “Each model has strengths! ChatGPT shines with creative tasks, Claude excels at processing large documents, and Gemini is strong with multimodal understanding.”
Another user remarked, “Each model has strengths! ChatGPT shines with creative tasks, Claude excels at processing large documents, and Gemini is strong with multimodal understanding."
A third user stated, “Why would someone need AI to answer this?”
A fourth comment read, “Thats why Claude is best for solving problems and critical thinking. Chatgpt can do the talking and writing content.”
A fifth user stated, “Must be an old version of Gemini & chatgpt. Recent ones are good.”
Fareha Naaz is a Delhi-based journalist and Content Producer at LiveMint, where she has built nearly three years of experience in digital journalism. She covers a diverse range of topics, including national news, education, entertainment, lifestyle trends science, global health and international news.<br><br> With a background in Economics and Education, she focuses on providing insightful, thoroughly researched coverage that bridges the gap between breaking news and in-depth analysis. In addition to breaking copies, legal and political news, her reporting blends editorial rigour with search-driven storytelling. With a keen eye-on-global events, she provides insightful coverage on latest developments. Her reporting combines editorial rigour with in-depth coverage and search-driven storytelling provide valuable insight and context to readers, ensuring accuracy and relevance.<br><br> Her newsroom experience helped her in combining her critical thinking skills with real-time editorial decision-making. Over the years, she has been presenting complex stories with clarity for a digital-first audience amid fast-paced news cycles. Her thoroughly researched stories, with well-structured and engaging content, provide readers with clear understanding of the context and background.<br><br> Fareha holds a Master’s degree in Economics, in addition to a Bachelor of Education degree.<br><br> When not in the newsroom, she enjoys painting and sports, reading books and current developments.