DeepSeek AI has gotten ChatGPT all worried. With its affordable large language models that it says can outperform competitors from industry titans like OpenAI and Meta, DeepSeek is upending the AI landscape. While the recently launched Janus Pro multimodal AI model is said to be able to outperform Stable Diffusion and DALL-E 3, the Chinese startup claims that its flagship R1 reasoning model can achieve “performance comparable” to OpenAI’s O1 counterpart.
DeepSeek AI, Upending Financial Markets
The company is upending financial markets as DeepSeek’s ChatGPT rival swiftly rose to the top of the App Store. On January 27, Nvidia’s stock fell 17%, removing almost $600 billion from its market capitalization, which CNBC reported was the worst single-day decline in US history.
DeepSeek Ai, Powered By DeepSeek-V3 model
The startup’s “state-of-the-art” DeepSeek-V3 model powers the AI assistant, which enables users to create text, plan trips, ask queries, and more. Due to “malicious attacks,” DeepSeek started limiting signups as app downloads surged.
DeepSeek Ai, Limiting Sign-Ups
In response to a surge in malicious cyberattacks targeting its innovative platform, DeepSeek AI Limited has taken the decisive step to limit new sign-ups temporarily. This proactive measure ensures the integrity and security of their cutting-edge AI tools, safeguarding the experience for current users. By reinforcing its defenses and thoroughly vetting new applicants, DeepSeek AI is committed to creating a secure and reliable environment where creativity and technological advancement can flourish. While this temporary limitation might feel like a setback, it ultimately serves to protect the community, fortifying the platform against any threats and paving the way for a future where trust and innovation go hand in hand.
READ ALSO: Vodafone Satellite Video Call, The First Of Its Kind
DeepSeek AI, Fewer GPUs, Lowering Costs
When compared to the billions spent by OpenAI, Meta, Google, Microsoft, and others, DeepSeek, which was founded in 2023 by Liang Wenfeng, has gained recognition for creating open-source AI models with fewer resources and GPUs. If DeepSeek’s performance claims are accurate, it may demonstrate that the startup was able to create potent AI models in spite of stringent US export regulations that forbade Nvidia and other chipmakers from selling high-end graphics cards in China.
DeepSeek Ai, Why Everyone is Freaking Out
By establishing more precisely defined expert categories and a more effective means of communication between them, the DeepSeek version introduced a novel idea that improved the efficiency of the training procedure. By compressing the way the model stores and retrieves data, the DeepSeek team also created DeepSeekMLA (Multi-Head Latent Attention), which significantly decreased the amount of memory needed to run AI models.
The world is shocked not only by the architecture that produced these models but also by how quickly it was able to duplicate OpenAI’s accomplishments in a matter of months as opposed to the year or more that usually separates significant advancements in AI, Brundage continued.
DeepSeek Ai, Open Ai
Due to its public image of being uniquely able to create advanced AI, OpenAI has recently gained the backing of investors to construct the largest AI data center infrastructure in the world. However, DeepSeek’s rapid replication demonstrates that even when businesses attempt to keep their techniques a secret, technical benefits are short-lived.
How Nvidia has Benefited From Ai hype cycles
Nvidia, a publicly traded corporation that manufactures the advanced chips used by AI businesses, has profited the most from the hype cycle. Purchasing Nvidia stock was thought to be an investment in the firm that was producing the shovels during the AI gold rush. To operate the models, whoever won the AI contest would require a large supply of Nvidia chips. The shares closed at $137.01 on December 27th, which is about ten times the value of Nvidia stock at the start of January 2023.
DeepSeek AI: A Threat to Nvidia Chip Prices
The investment theory that propelled Nvidia to exorbitant pricing is overturned by DeepSeek’s performance. Other businesses will follow suit if the corporation is actually employing chips more effectively rather than just purchasing more of them. As businesses attempt to reduce their expenditures, this could result in a smaller market for Nvidia’s most cutting-edge chips.