Plain-English technology

Useful technology, explained without the noise.

Life Me Tech publishes practical guides for AI, phones, apps, online safety and the digital tools you use every day.

Explore latest guidesHow we work
Showing posts with label Chatbots. Show all posts
Showing posts with label Chatbots. Show all posts

Sunday, 18 October 2026

I Asked ChatGPT, Gemini and Claude the Same 7 Questions — One Was Clearly Better

Everyone has a favourite AI chatbot – and everyone is convinced theirs is the smartest. So I stopped arguing and ran a proper test instead: the same seven real-world tasks, fed to ChatGPT, Google Gemini and Anthropic’s Claude, back to back. The results genuinely changed which one I open first every morning.



Test 1: Writing a Professional Email

I asked each bot to draft a follow-up email after a job interview – polite, confident, under 150 words. All three produced usable drafts in seconds, but Claude’s stood out: it sounded like an actual human wrote it, not a template. ChatGPT’s version was solid but a touch stiff. Gemini’s was the shortest, which is fine if you like editing down rather than up. Winner here: Claude, by a clear margin.

Test 2: Summarising a Long Article

Next I pasted in a 4,000-word article about the housing market and asked for five bullet points. This is where Gemini shone – its summary caught two nuances the others missed, and it offered to turn the summary into a table without being asked. ChatGPT did a decent job but skipped the article’s main criticism. Claude’s summary was accurate but oddly long. If you summarise documents all day, Gemini earned its place.

Test 3: Fixing a Bit of Broken Code

I’m no programmer, so I grabbed a small broken Python script from a beginner tutorial and asked each bot to find the bug. ChatGPT found it instantly and – more usefully – explained why it broke in plain English. Claude also found it but buried the answer in explanation. Gemini spotted the bug too, then confidently suggested a "fix" that introduced a new error. For coding help, ChatGPT remains the one to beat.

Test 4: Planning a Weekend Trip

I asked for a two-day Austin itinerary on a $300 budget. Gemini pulled ahead thanks to its Google Maps integration – it mapped the route between stops and flagged which restaurants were actually open on Sundays. ChatGPT gave a charming plan but listed one restaurant that closed in 2024. Claude’s itinerary was thoughtful but ignored my budget entirely. For anything location-based, Gemini’s live data is a genuine superpower.

Test 5: A Tricky Logic Question

Each bot got the same multi-step maths puzzle. Claude solved it correctly and showed its working step by step, which made it easy to trust. ChatGPT got the right answer but skipped a step, so I had to ask it to explain itself. Gemini got the answer wrong – twice – and sounded very confident both times. Lesson learned: never trust an AI’s maths without checking, especially Gemini’s.

Test 6: Brainstorming Business Ideas

I asked for ten side-hustle ideas requiring less than $100 to start. ChatGPT’s list was practical and specific – it even estimated startup costs. Claude’s ideas were more creative but half of them needed way more than $100. Gemini gave generic ideas I’d already seen on a hundred listicles. For creative brainstorming with realistic detail, ChatGPT took this round.

Test 7: Following Complicated Instructions

The final test was deliberately fussy: rewrite a paragraph in a casual tone, keep it under 80 words, use exactly three bullet points, and end with a question. Claude nailed every constraint on the first try. ChatGPT hit most of them but used four bullets. Gemini ignored the word limit entirely. When precision matters – legal-ish drafts, formatting rules, strict briefs – Claude is the most obedient of the three.

Quick win: stop picking one chatbot for everything. Use ChatGPT for coding and brainstorming, Gemini for anything involving maps, places or current events, and Claude for writing and precise instructions. Bookmark all three free versions – switching between them takes seconds and the quality jump is real.

So who won overall? Honestly, it depends on your day. But if I had to pick one daily driver, it’s Claude – its writing quality and instruction-following won the tests I run most often. ChatGPT is the best all-rounder, and Gemini is unbeatable when live information matters. The real takeaway: the best AI assistant in 2026 isn’t one app, it’s knowing which one to open for which job.

Want to build your own online business for free? Start here

Life Me Tech