An AI training gap<\/a> has emerged across the business world. According to one recent survey, two-thirds of leaders expect employees to have AI skills, but only a third of companies have clear policy on what technology to use\u2014and how to use it.<\/p>\n While utilizing generative AI chatbots has a relatively low barrier to entry, figuring out the right <\/em>model remains a challenge for people who have not studied their nuances. Here\u2019s what you need to know about picking the right AI tools, such as ChatGPT<\/a>, Claude<\/a>, Gemini<\/a>, Perplexity<\/a>, and Copilot<\/a>\u2014and the underlying models that power them.<\/p>\n Learn more: <\/strong>Mastering AI at work: a practical guide to using ChatGPT, Gemini, Claude, and more<\/a><\/p>\n Today\u2019s generative AI is still relatively new, and for this reason, there is an abundance of models on the market; no one company or model dominates. As a result, experimenting with a variety of tools is beneficial.<\/p>\n \u201cThink of it like test driving a car,\u201d Jules White, a computer-science professor at Vanderbilt University, told Fortune<\/em>. \u201cWhat is it like to drive on the highway, to park, how does the stereo work, how cushy is the seat, etc. I suggest doing the equivalent for models.\u201d<\/p>\n For Maggie Vo, head of user education at Anthropic, that test drive should include consideration of three factors: task complexity, time sensitivity, and the need to refine your work.\u00a0<\/p>\n Something like writing a strategic plan might call for using the most capable model whereas a faster model might be most useful for reformatting data or quick summaries. If you\u2019re planning to iterate multiple times, then use a combination: \u201cStart with a smarter model to refine your approach, then use a smaller model to execute,\u201d she said.<\/p>\n \u201cThe real skill is developing \u2018Platform Awareness\u2019\u2014understanding not just different models but different AI systems entirely,\u201d Vo added. \u201cWhat works in Claude might need adjustment in other systems. Experiment across platforms to build intuition about their unique strengths.\u201d<\/p>\n In practice, this can be as simple as typing the same prompt in different AI chatbots\u2014as well as the models for each one (the option to change may be at top of the screen or close to the search bar)\u2014and compare which tool gives the most helpful response.<\/p>\n Screenshot from Anthropic\u2019s Claude<\/p>\n<\/div>\n As Vo implied, as AI models have been developed, certain tasks are more easily completed with some AI models over others. The faster, typically free models, are usually best for simple chats\u2014whereas the more advanced models may cost a fee and take more than a few moments to process, but will typically yield better results.<\/p>\n Ethan Mollick<\/a>, a professor at the University of Pennsylvania\u2019s Wharton School\u2014is one of the most widely-read experts on how to use AI models to help with business tasks. He is known for his prolific research and analysis on LinkedIn<\/a>. According to a recent Substack post<\/a>, his everyday AI use is focused within the products of OpenAI, Anthropic, and Google.<\/p>\n \u201cWith all (three) of the options, you get access to both advanced and fast models, a voice mode, the ability to see images and documents, the ability to execute code, good mobile apps, the ability to create images and video (Claude lacks here, however), and the ability to do Deep Research,\u201d he wrote.<\/p>\n The ultimate challenge comes with deciding which model is best to use, he added. But like White, he connects it to selecting a vehicle: \u201cThink of it like choosing between a sports car and a pickup truck; both are vehicles, but you\u2019d use them for very different tasks.\u201d<\/p>\n Mollick summarized the decision making into three categories: good for chats, good for work, and good for hard problems.<\/p>\n Claude 4 Sonnett, ChatGPT-4o, and Gemini 2.5 Flash are fast\u2014and not as smart, according to Mollick. But as generally free AI products, they are the most widely accessible and can still make a difference in workplace efficiency.<\/p>\n Here are some examples of their capabilities:<\/p>\n Claude 4 Opus, ChatGPT o3, and Gemini 2.5 Pro are smarter, have lower error rates, but are slower, according to Mollick. They may still hallucinate, but tend to be more successful at coding, logic, or solving tougher problems that require multiple steps. Here\u2019s an example of their capability:\u00a0<\/p>\n Who better to summarize AI\u2019s priorities than AI itself? Here\u2019s what the models had to say:<\/p>\n Prompt<\/strong>: \u201cIn one sentence, please describe your relative strengths. I am writing this in the context of other generative tools, including [Competitor] and [Competitor].\u201d<\/p>\n OpenAI\u2019s GPT-4o:<\/strong> \u201cChatGPT\u2019s relative strength lies in its ability to generate highly polished, context-aware, and conversationally natural responses\u2014particularly strong in editing, writing, and task-oriented problem-solving\u2014while integrating tools like code execution and image analysis more seamlessly than Claude or Gemini.\u201d<\/p>\n Anthropic\u2019s Claude Sonnet 4:<\/strong> \u201cI excel at nuanced reasoning, ethical decision-making, and producing well-structured, thoughtful responses while maintaining strong safety guardrails, though I\u2019m generally more cautious and less willing to engage with borderline requests compared to some other AI assistants.\u201d<\/p>\n Google\u2019s Gemini 2.5 Flash:<\/strong> \u201cGemini excels in multimodal understanding and generation, seamlessly integrating text, code, images, and other data formats while leveraging real-time information from Google\u2019s vast ecosystem to provide highly accurate and contextually rich responses.\u201d<\/p>\n Some other examples include the ability to conduct customer-service automation & quality control as well as financial-scenario modeling & risk analysis. Mollick admitted he primarily uses the \u201cGood for work\u201d models, though he also has favorites for specific tasks outside of this list, like using GPT-4.5 when writing.<\/p>\n Claude 4 Opus extended thinking, ChatGPT-o3-pro, and Gemini 2.5 Pro are the most advanced AI models on the market, and for that reason, they remain slow and have limited use, Mollick added in his Substack post.<\/p>\n Here\u2019s an example of a prompt that would push the limit of ChatGPT of, but ChatGPT-o3-pro might be best equipped to tackle:<\/p>\n \u201cBelow is the full text of a 180-page Phase III oncology trial dossier (including statistical appendices and 40 pages of adverse-event tables).<\/p>\n No matter what model or AI company used, it\u2019s always important to double check the AI for errors or hallucinations. Accuracy is the priority of Perplexity, according to Jesse Dwyer<\/a>, head of communications at the AI company.<\/p>\n \u201cPerplexity\u2019s sole focus is accurate, trustworthy AI\u2014we use all top models and post-train them for accuracy,\u201d Dwyer said. \u201cModels that have been trained to experience some hallucination are helpful if you want videos of high-diving cats, but they can be dangerous when you are making business or financial decisions.\u201d<\/p>\n Copilot is also a widely used AI chatbot, thanks to its integration with Microsoft<\/a> products, but it can be difficult to switch between models. DeepSeek r1 and Grok by Elon Musk\u2019s xAI also are options on the market, but, according to Mollick, each have missing features.<\/p>\n Mollick did not respond to Fortune<\/em>\u2019s request for comment.<\/p>\n While there remains no perfect method for using AI in the workplace, the most effective ways to remain on top of your game is exploration and education\u2014and that starts with simply using them.<\/p>\n \u201cThe difference between casual users and power users isn\u2019t prompting skill (that comes with experience); it\u2019s knowing these features exist and using them on real work,\u201d Mollick wrote.<\/p>\n And because many business leaders, including CEOs, have already begun using AI, Dwyer suggested trying to emulate how they are using it in their work.<\/p>\n \u201cAI is one of the first business tools adopted first by managers instead of frontline workers,\u201d he told Fortune<\/em>. \u201cIt makes sense that leaders with experience getting the best work from their teams and software tools will be naturally prepared to work with AI.\u201d<\/p>\n<\/div>\n Introducing the 2025 Fortune 500<\/strong>, the definitive ranking of the biggest companies in America.\u00a0Explore this year’s list.<\/a><\/p>\nSpeed date with AI<\/h2>\n
![](\"https:\/\/fortune.com\/img-assets\/wp-content\/uploads\/2025\/07\/image.png?w=1440&q=85\")
<\/p>\nThe best models, according to UPenn\u2019s Ethan Mollick<\/h2>\n
Good for chats<\/h3>\n
\n
Good for work\u00a0<\/h3>\n
\n
![](\"https:\/\/fortune.com\/img-assets\/wp-content\/uploads\/2025\/07\/AI-Screenshot.png?w=1440&q=85\")
<\/p>\nGood for hard problems<\/h3>\n
\n
Other models<\/h3>\n
The takeaway: Practice makes perfect<\/h2>\n