{"id":13950,"date":"2025-01-28T02:50:48","date_gmt":"2025-01-28T02:50:48","guid":{"rendered":"https:\/\/topat10.com\/?p=13950"},"modified":"2025-01-28T02:50:48","modified_gmt":"2025-01-28T02:50:48","slug":"deepseek-vs-chatgpt-hands-on-with-deepseeks-r1-chatbot","status":"publish","type":"post","link":"https:\/\/topat10.com\/?p=13950","title":{"rendered":"DeepSeek vs. ChatGPT: Hands On With DeepSeek\u2019s R1 Chatbot"},"content":{"rendered":"


\n<\/p>\n

\n

The DeepSeek AI<\/span> chatbot, released by a Chinese startup, has temporarily dethroned<\/a> OpenAI\u2019s ChatGPT from the top spot on Apple\u2019s US App Store.<\/p>\n

The app is completely free to use, and DeepSeek\u2019s R1 model is powerful enough to be comparable to OpenAI\u2019s o1 \u201creasoning\u201d model, except DeepSeek\u2019s chatbot is not sequestered behind a $20-a-month paywall like OpenAI\u2019s is. Also, the DeepSeek model<\/a> was efficiently trained using less powerful AI chips, making it a benchmark of innovative engineering.<\/p>\n

I\u2019ve tested many new generative AI tools<\/a> over the past couple of years, so I was curious to see how DeepSeek compares to the ChatGPT<\/a> app already on my smartphone. After a few hours of using it, my initial impressions are that DeepSeek\u2019s R1 model will be a major disruptor for US-based AI companies, but it still suffers from the weaknesses common to other generative AI tools, like rampant hallucinations, invasive moderation, and questionably scraped material<\/a>.<\/p>\n

How to Access the DeepSeek Chatbot<\/h2>\n

Users interested in trying out DeepSeek can access the R1 model through the Chinese startup\u2019s smartphone apps (Android<\/a>, Apple<\/a>), as well as on the company\u2019s desktop website<\/a>. You can also use the model through third-party services like Perplexity Pro. In the app or on the website, click on the DeepThink (R1)<\/strong> button to use the best model. Developers who want to experiment with the API can check out that platform<\/a> online. It\u2019s also possible to download a DeepSeek model to run locally<\/a> on your computer.<\/p>\n

In order to use all the consumer features, you will need to create a user account that tracks your chats. \u201cWe store the information we collect in secure servers located in the People’s Republic of China,\u201d reads the company\u2019s privacy policy. Check out this article<\/a> from WIRED\u2019s Security desk for a more detailed breakdown about what DeepSeek does with the data it collects. It\u2019s worth keeping in mind that, just like ChatGPT and other American chatbots, you should always avoid sharing highly personal details or sensitive information during your interactions with a generative AI tool.<\/p>\n

Is This Basically FreeGPT?<\/h2>\n

Yes and no! If you\u2019re looking for a free chatbot to use, ChatGPT already includes plenty of free features. So does Anthropic\u2019s Claude, Google\u2019s Gemini, and Meta\u2019s AI tool. So, why is the fact that DeepSeek is free notable? It\u2019s about the raw power of the model that\u2019s generating these free-for-now answers. As previously mentioned, DeepSeek\u2019s R1 mimics OpenAI\u2019s latest o1 model, without the $20-a-month subscription fee for the basic version and $200-a-month for the most capable model. This comes as a major blow to OpenAI\u2019s attempt to monetize ChatGPT through subscriptions.<\/p>\n

Another feature that\u2019s similar to ChatGPT is the option to send the chatbot out into the web to gather links that inform its answers. DeepSeek does not have deals with publishers to use their content in answers; OpenAI does , including with WIRED\u2019s parent company, Cond\u00e9 Nast. But the web search outputs were decent, and the links gathered by the bot were generally helpful.<\/p>\n

Still, the current DeepSeek app does not have all the tools longtime ChatGPT users may be accustomed to, like the memory feature<\/a> that recalls details from past conversations so you\u2019re not always repeating yourself. DeepSeek also doesn\u2019t have anything close to ChatGPT\u2019s Advanced Voice Mode<\/a>, which lets you have voice conversations with the chatbot, though the startup is working on more multimodal capabilities<\/a>.<\/p>\n

A Research Breakthrough, but Still Inaccurate<\/h2>\n

Though it may almost seem unfair to knock the DeepSeek chatbot for issues common across AI startups, it\u2019s worth dwelling on how a breakthrough in model training efficiency does not even come close to solving the roadblock of hallucinations, where a chatbot just makes things up in its responses to prompts. Many of the outputs I generated included blatant falsehoods, confidently spewed out. For example, when I asked R1 what the model already knew about me without searching the web, the bot was convinced I\u2019m a longtime tech reporter at The Verge. No shade, but not true!<\/p>\n

\n