DeepSeek is the name of a free AI-powered chatbot that has many of the same features as ChatGPT. But while \u201call ink is good ink,\u201d the publicity has been a two-edged sword for the controversial AI upstart.<\/strong><\/p>\n Like when any new AI model is released, many organizational leaders are asking, \u201cShould my company use DeepSeek?\u201d<\/p>\n Security and privacy issues may complicate the answer.<\/p>\n DeepSeek\u2019s models can collect and process data free from legal fetters. For instance, an AI model built by a company in the U.S. generally can\u2019t collect personal identification information (PII) without the user\u2019s consent. In contrast, those built abroad don\u2019t have to concern themselves with U.S. state laws.<\/p>\n DeepSeek\u2019s privacy policy<\/a> doesn\u2019t explicitly say they won\u2019t collect data without user consent. It merely says a user \u201cmay have a right to access, change, oppose \u2026 withdraw [consent], or limit\u201d how DeepSeek collects their data.<\/p>\n For some personal and business users, DeepSeek\u2019s vague consent policy may feel inadequate.<\/strong><\/p>\n From a data security perspective, many are having second thoughts about trusting DeepSeek. Their servers exist in China, which means the Chinese government \u2014 or the people it employs \u2014 has the power to easily access them. Given China\u2019s questionable record regarding human rights and privacy, some have expressed concern.<\/p>\n For instance, suppose you\u2019re a developer for a fintech company, and you\u2019re coding software designed to automatically manage sensitive user information. Instead of using ChatGPT or Gemini, you hop on DeepSeek to get help writing your code. The code DeepSeek produces now lives on a server in China. If the Chinese government wanted to hack the software you created, they could simply hand the code to a software engineer and say, \u201cPlease find the vulnerabilities in this code.\u201d<\/p>\n On a larger scale, a company may choose to use DeepSeek to build a code generation tool, similar to solutions provided by companies such as Replit<\/a> and Qodo<\/a>. Every user who turns to the solution to build code would have their programming stored on a server in China, which may introduce more risk than a programmer is willing to accept.<\/p>\n DeepSeek\u2019s model has built-in censorship mechanisms, such as those that eliminate content around controversial topics like the Tiananmen Square killings<\/a> or the Chinese treatment of Uyghurs. It also puts a pro-China and anti-U.S. spin on the content it produces with \u201cSorry, that\u2019s beyond my current scope. Let\u2019s talk about something else,\u201d a common refrain for politically sensitive topics.<\/p>\n The model can change or suppress information that doesn\u2019t align with its backers’ political leanings.<\/p>\n Beyond censorship, Chinese manufacturers have a long history of copying U.S. designs, producing similar products, and then selling them for far less money. DeepSeek not only puts China in a leadership position in the AI market but also enables AI developers with shallower pockets to create products at a fraction of the cost they\u2019d pay OpenAI or other U.S. companies.<\/strong><\/p>\n While they may seem similar in terms of capabilities, a few key differences can help leaders better evaluate DeepSeek.<\/p>\n DeepSeek claims its AI model is as powerful as OpenAI\u2019s models, yet they reportedly built it for only $5.6 million<\/a>. While this is being refuted<\/a>, the announcement had a ripple effect that felt like a tidal wave. NVIDIA\u2019s stock and other tech heavyweights dipped<\/a> as news of a cheaper way of producing AI splashed across the ticker.<\/p>\n Let\u2019s compare the models.<\/p>\n The training method is one of the biggest differences between DeepSeek and OpenAI. While OpenAI’s o1 models<\/a> use large-scale Supervised Fine-Tuning (SFT) combined with reinforcement learning, DeepSeek started R1-Zero using only reinforcement learning \u2014 a first for open-source models.<\/p>\n On the other end of the AI-training spectrum is a dense or monolithic model. With this approach, a huge stack of computational resources joins forces to solve each problem in the training process. If you think of each computational unit as a node in a network, imagine every single node firing up simultaneously and all of them working together to train a single facet of an AI model. That process consumes much more energy.<\/p>\n Another differentiator is that DeepSeek used the mixture of experts (MoE) model to build its solution, which may have resulted in lower computational costs. Unlike traditional AI models, where every part processes all tasks, MoE divides the model into specialized sub-networks, each focusing on specific problems. A routing system then directs incoming tasks to the most appropriate expert, ensuring efficient processing.<\/p>\n In other words, a model specifically designed to handle a certain type of AI training gets assigned to do what it does best.<\/strong> Only this single \u201cexpert\u201d must expend energy toward building the element of the model. That saves money.<\/p>\n To illustrate, imagine you have a team of specialists ready to build 10 houses. Some are electricians, others specialize in flooring, others are plumbers, some are framers, and so on. You tell all of them to start working on the framing.<\/p>\n Even though they all try their best, the electricians and plumbers expend far more energy \u2014 and make far more mistakes \u2014 while framing walls than those who specialize in the role. Similarly, the plumbers, even if they try their best, may struggle with installing three-way switches, choosing the right size breakers, and other electrician-specific tasks.<\/p>\n On the other hand, suppose you assign your framers, plumbers, electricians, flooring specialists, and others to teams. You then tell them to tackle only the tasks they specialize in. They would expend far less energy and complete the job much faster. As the general contractor, you would save considerable cash along the way.<\/p>\n By using an MoE approach, DeepSeek did the same thing. Thus, they produced a high-quality product more efficiently.<\/p>\n There\u2019s a good chance DeepSeek obtained high-quality data at a fraction of the cost of other AI leaders, and OpenAI says it has proof DeepSeek used OpenAI data to train its model. Granted, no one knows exactly how much OpenAI and others have paid for the rights to use data from encyclopedias, newspapers, medical journals, and more, but the price tag may have been significant.<\/p>\n DeepSeek may have gained access to equally robust datasets for very little money. For example, since the company has the Chinese government\u2019s support, it could\u2019ve had a government agency foot the bill for costly data. Then, it could run its MoE models on data that cost little to nothing.<\/p>\n DeepSeek may be getting significant financial support from the Chinese government. For a private sector organization, such as OpenAI and other leading tech companies, it can be tough to secure government funding because of the fear of taxpayer backlash.<\/p>\nPotential Risks Associated with DeepSeek<\/h2>\n
Censorship and Political Bias Issues<\/h3>\n
What\u2019s the Big Difference Between DeepSeek and Other Models?<\/h2>\n
Deepseek\u2019s Training Method<\/h3>\n
Mixture of Experts Model<\/h3>\n
The Cost of Data<\/h3>\n
Government Support<\/h3>\n
Reinforcement Learning<\/h3>\n