On February 21, 2025, Beijing time, the DeepSeek AI team, focused on exploring Artificial General Intelligence (AGI), made an exciting announcement on social media, officially kicking off their participation in the #OpenSourceWeek event. This small yet ambitious tech team announced that starting next week, they will open source five code repositories to share their research progress transparently with the global developer community.

Recently, the Evo2 bio AI model, jointly developed by Arc Institute and Nvidia, was officially released. This foundational model is based on DNA data from over 100,000 species and aims to deeply decode various complex phenomena in biology. Evo2 can identify patterns in the genetic sequences of different organisms that researchers would typically take years to discover, significantly enhancing the ability to identify disease-related mutations, and can design entirely new genomes comparable to simple bacteria.

Recently, AI programming startup Codeium is undergoing a new funding round, with a valuation close to $2.85 billion. According to sources, this funding round is led by investment firm Kleiner Perkins, marking another occasion where Codeium has garnered investor attention after six months. Notably, in the past six months, Codeium completed a $150 million Series C funding round, with a post-funding valuation of $1.25 billion. Although the specific amount for the new funding round has not been disclosed yet.

Recently, an open-source plugin that seamlessly integrates the DeepSeek R1 model into Visual Studio Code has gained widespread attention in the developer community, with installation numbers in the VSCode plugin market surpassing 40,000. This plugin provides developers with a one-stop AI programming assistant service.

Recently, cybersecurity researchers discovered two malicious machine learning models that were uploaded quietly to the renowned machine learning platform HuggingFace. These models utilized a novel technique, successfully bypassing security detection through 'corrupted' pickle files, raising concerns. Karlo Zanki, a researcher at ReversingLabs, pointed out that the beginning of the pickle files extracted from these PyTorch format archives suggests that

DingTalk has announced that its AI Assistant has fully integrated with the DeepSeek model series, providing users with more intelligent options. When creating an AI Assistant, users can choose from the DeepSeek R1, V3, and various distilled models. Additionally, DingTalk has launched a brand-new template that simplifies the AI Assistant creation process, enabling users to publish assistants based on the DeepSeek models with just one click. The AI Assistant based on DeepSeek not only possesses deep thinking capabilities but also supports online queries.

Recently, 59 participants from different backgrounds gathered at the AI2Web3 Bootcamp held in New York to explore how to leverage these two cutting-edge technologies to create practical products and services. The bootcamp was hosted by Matheus Pagani, founder and CEO of Venture Miner, who pointed out that understanding the core components of these two technologies will enable people from all sectors to participate in future innovations. During the bootcamp, Pagani demonstrated how AI can be showcased through simple writing exercises.

Recently, Shanghai Jumpsky Star Intelligent Technology Co., Ltd.'s Jumpsky App launched a new feature called 'AI Creative Board', providing users with an innovative experience to develop applications without coding skills. In the digital era, many creative ideas struggle to materialize due to a lack of programming skills. The 'AI Creative Board' feature aims to break this barrier, lowering the threshold for application development to its minimum. Users simply need to upgrade the Jumpsky App to the latest version, click on the 'Creative Board' at the bottom left, and use simple text descriptions (Prompt

A research team from the United States and the United Kingdom has recently succeeded in reviving a chatbot code known as ELIZA, which is considered the first electronic chatbot in history. According to their paper published on the arXiv preprint server, this code was originally written by the late MIT professor Joseph Weizenbaum in the 1960s. In 2021, an MIT archivist named Miles Crowley discovered printed copies of this code in a box containing Weizenbaum's personal belongings. The revival of this code marks an important milestone in the history of artificial intelligence.

In recent years, the programming capabilities of Artificial Intelligence (AI) have been continuously evolving, but they have not yet reached perfection. Recently, Max Woolf, a senior data scientist at BuzzFeed, discovered through experiments that if continuous prompts for 'writing better code' are provided to large language models (LLMs), AI can indeed generate higher quality code. This finding has garnered widespread attention, with renowned AI scientists emphasizing the importance of iteration and prompt design. In Woolf's experiment,

Recently, Max Wolf, a senior data scientist at BuzzFeed, conducted an experiment exploring the effects of repeatedly asking AI to improve code. In the experiment, he used the Claude3.5 language model and presented a classic programming challenge: to write Python code to find the difference between the maximum and minimum values of numbers that sum to 30 among one million random numbers. Image Source Note: Image generated by AI, with initial version authorized by Midjourney.