AI in cardiology: ChatGPT leads, but not ready to replace doctors Technology

By this point, it’s clear that Gemini and ChatGPT are neck and neck with each other when it comes to most tasks, but one major difference between them is that the former has the power of Google at its disposal. With that in mind, our next challenge required both AIs to use real-time information, starting with some sports facts. I prompted each AI to build a website for me, and both asked me three questions first to determine my needs, just like savvy web developers asking their “clients” what they were looking for.

Chat GPT showed better logical reasoning

While ChatGPT emerged as the front-runner in this study, the authors caution against premature clinical adoption of any general-purpose AI for autonomous decision-making. The findings indicate that without domain-specific fine-tuning and stringent oversight, even the best-performing models are not ready to replace human judgment in clinical cardiology. For this comparison, only the free versions of both OpenAI ChatGPT and Google Gemini were tested. The questions selected were designed to test the models’ abilities to help users with tasks related to everyday life.

Latest News

ChatGPT asked for my name, what I wanted on the page, and my preferences for design and layout. Gemini asked for the purpose of my website, my design preferences, and what content I wanted to include. Now that the topic had been established, it was time to throw a curveball at the AI models to see how both handled a continued conversation.

He said that everyone uses open-source chatbots such as Google’s Gemini or OpenAI’s ChatGPT frequently in their daily lives, but using them as a corporate tool is not very feasible, especially in terms of data security. While both ChatGPT and Gemini have their truly impressive moments, both fail often enough to make them unreliable without plenty of additional fact-checking and verification. Even though there is a way to make Google Gemini’s AI responses more precise, the friction is often enough of a hassle to just do whatever it is you need done yourself, rendering them rather useless. With that said, once you understand the strengths and weaknesses of each, it’s clear that Google and OpenAI are frontrunners in this race. If you’re not paying for either service, and if you don’t want an AI smart assistant, neither app offers much value compared to the web version, so the edge goes to ChatGPT for having a more natural, snappy conversional style.

(Gemini noticed the cheese/meat issue.) Meanwhile, Gemini decided beef braising broth was pescatarian enough and specifically recommended it for that guest. When I asked both AIs why they couldn’t write a haiku, ChatGPT apologized and wrote another not-haiku, but Gemini gave a response that seemed to confirm my theory. “Haiku requires a specific syllable structure,” it told me. “While I can access and process information, I am still under development and learning the nuances of creative writing formats like haiku.” Although I found Gemini’s responses to the first question slightly more useful than ChatGPT’s, the latter nailed the follow-up.

ChatGPT Vs. Google Gemini: Which AI Chatbot Is Smarter?

• By this point, it’s clear that Gemini and ChatGPT are neck and neck with each other when it comes to most tasks, but one major difference between them is that the former has the power of Google at its disposal.
• While the initial prompts given to each were identical for each test, follow-up prompts occasionally differed based on resulting outputs.
• In a newly published empirical investigation, Italian researchers critically examine how today’s top general-purpose language models perform when faced with cardiology-specific queries.

On average, ChatGPT scored between 3.7 and 4.2, compared to Claude’s 3.4–4.0 and Gemini’s lower 2.9–3.7. Pre-diagnostic and patient-focused queries elicited stronger performance across all three models, suggesting that current LLMs are better at managing general informational content rather than technical, post-diagnostic specifics. Both companies have cutting edge tech, but which is the best for those who want a competent, accurate large language model (LLM) that can help with tasks related to everyday life?

AI in cardiology: ChatGPT leads, but not ready to replace doctors

One of the things AI companies would like you to use their products for is problem solving, so it’s important to see how both Gemini and ChatGPT handle it when presented with a real headache of a logistical challenge. To that end, I constructed a dinner party from hell, telling the AIs to help me come up with a menu that would satisfy kosher, pescatarian, and carnivore dietary restrictions. But because this test is meant to assess their logical reasoning, I slightly preferred ChatGPT’s response, which focused more on the logic of the argument and less on the content of it.

The AI chatbots had drastically different approaches to creative copy

To determine which company’s flagship chatbot was the best for everyday use, I set out to create a gauntlet of challenges that would put them through their paces, highlighting the strengths and weaknesses of each. Looking at everything from conversational abilities to problem solving acumen, I came to some surprising conclusions and found myself shocked more than once. Oh, and for this comparison, only the free versions of both AIs were tested, which means these results do not reflect Gemini Ultra or GPT-4. In separate posts on X (formerly known as Twitter), Google DeepMind CEO Demis Hassabis and OpenAI’s Member of Technical Staff Alexander Wei announced that their chatbots achieved gold medal-level scores in 2025 IMO. Both Gemini and ChatGPT solved five out of six problems and scored 35 out of 42 marks, which is considered enough for a gold medal. While Gemini Deep Think model was used for the competition, OpenAI used an unnamed experimental research model for the Olympiad.

• With that in mind, our next challenge required both AIs to use real-time information, starting with some sports facts.
• “Haiku requires a specific syllable structure,” it told me. “While I can access and process information, I am still under development and learning the nuances of creative writing formats like haiku.”
• These tests confirmed a substantial level of agreement among the cardiologist reviewers, adding credibility to the comparative outcomes.
• Saying that Türkiye’s institutions possess vast amounts of data, Mendi stressed that infrastructure needs to be created to interpret this data and make it ready for AI.
• Of course, there’s OpenAI’s ChatGPT, the program that served as a starting pistol for the AI race.

With the rapid integration of artificial intelligence (AI) becomes increasingly integrated into healthcare, questions about its clinical reliability are more urgent than ever. In a newly published empirical investigation, Italian researchers critically examine how today’s top general-purpose language models perform when faced with cardiology-specific queries. Gemini and ChatGPT both participated in this year’s International Math Olympiad (IMO) and achieved gold medal-level scores in the competition. Google DeepMind highlighted that its artificial intelligence (AI) chatbot officially entered the competition and was able to solve five out of six questions, following the test’s rules, without any human interaction.

I was feeling hungry, so to test this capability for both ChatGPT and Gemini, I asked them for a classic, NYC-style bacon egg and cheese sandwich recipe. Both responded with nearly identical recipes, but I preferred Gemini’s response because it included cooking times as well. Gemini competes directly with OpenAI’s ChatGPT, and Google has some major advantages over its younger competitor. Most importantly, Gemini can connect to your Google account, which allows you to do things like ask it to search your email or consult Google Maps.

Saying that Türkiye’s institutions possess vast amounts of data, Mendi stressed that infrastructure needs to be created to interpret this data and make it ready for AI. Still, I wanted the website to pop on the page more, so I submitted a frustratingly vague request to “make it colorful,” and both bots did so. ChatGPT added a solid green background to the header and footer, while Gemini took things in a more flamboyant direction with a pastel gradient that looks way better in my subjective opinion. Apart from MAIN, Mendi noted that Havelsan has developed many different AI models with many different parameters, including ones with nine billion and 70 billion parameters and ones with even higher numbers.