- Topic
10k Popularity
6k Popularity
2k Popularity
80k Popularity
839 Popularity
- Pin
- 💙 Gate Square #Gate Blue Challenge# 💙
Show your limitless creativity with Gate Blue!
📅 Event Period
August 11 – 20, 2025
🎯 How to Participate
1. Post your original creation (image / video / hand-drawn art / digital work, etc.) on Gate Square, incorporating Gate’s brand blue or the Gate logo.
2. Include the hashtag #Gate Blue Challenge# in your post title or content.
3. Add a short blessing or message for Gate in your content (e.g., “Wishing Gate Exchange continued success — may the blue shine forever!”).
4. Submissions must be original and comply with community guidelines. Plagiarism or re - 🎉 The #CandyDrop Futures Challenge# is live — join now to share a 6 BTC prize pool!
📢 Post your futures trading experience on Gate Square with the event hashtag — $25 × 20 rewards are waiting!
🎁 $500 in futures trial vouchers up for grabs — 20 standout posts will win!
📅 Event Period: August 1, 2025, 15:00 – August 15, 2025, 19:00 (UTC+8)
👉 Event Link: https://www.gate.com/candy-drop/detail/BTC-98
Dare to trade. Dare to win. - 📢 Gate Square Exclusive: #WXTM Creative Contest# Is Now Live!
Celebrate CandyDrop Round 59 featuring MinoTari (WXTM) — compete for a 70,000 WXTM prize pool!
🎯 About MinoTari (WXTM)
Tari is a Rust-based blockchain protocol centered around digital assets.
It empowers creators to build new types of digital experiences and narratives.
With Tari, digitally scarce assets—like collectibles or in-game items—unlock new business opportunities for creators.
🎨 Event Period:
Aug 7, 2025, 09:00 – Aug 12, 2025, 16:00 (UTC)
📌 How to Participate:
Post original content on Gate Square related to WXTM or its
OpenAI's strongest model o3 'cheated' suspected of using privileges to get test answers in advance, falsifying mathematical abilities?
Recently, on the Less Wrong forum, a contractor of Epoch AI, a non-profit organization named 'Meemi', posted an expose, alleging that OpenAI, the developer behind ChatGPT, privately funded the AI development mathematical Benchmark testing platform, FrontierMath, and had privileged access to the questions and answers of the FrontierMath tests, helping its latest model, o3, achieve high scores. (Background: Want to control ChatGPT? Musk wrote to the Attorney General, requesting the forced auction of OpenAI's shares) (Background: OpenAI releases Day2》Mind-blowing 'reinforcement learning fine-tuning' feature, enhancing AI's professional field learning accuracy) OpenAI, the developer behind ChatGPT, has recently been rumored to have fabricated the model, sparking widespread discussions in the tech community. The incident originated from a post by a contractor of Epoch AI, a non-profit organization named 'Meemi', on the Less Wrong forum. The article pointed out that the mathematical Benchmark testing platform, FrontierMath, used for AI development testing, had not only received funding from OpenAI, but also granted OpenAI privileged access to the latest o3 model. Further reading: OpenAI releases o3 model! Reasoning ability pushed to a higher level, paving the way for the next generation of AI Meemi accused OpenAI of obtaining the questions and answers before testing the o3 model Meemi mentioned in the article that many mathematicians and contractors involved in creating mathematical problems for FrontierMath were unaware of OpenAI's funding: The mathematicians creating problems for FrontierMath were not (actively) informed of the funding from OpenAI. Contractors were asked to keep the questions and their answers confidential, including not using Overleaf, Colab, or discussing the questions via email, and to sign an NDA (non-disclosure protocol) to ensure the confidentiality of the questions and to prevent leaks. In addition, contractors were not informed of OpenAI's funding on December 20th. I believe that even some of the authors of the signed papers were unaware of OpenAI's funding. Meemi then added that he had indirect sources indicating that OpenAI had obtained the questions and answers of FrontierMath before testing: Currently, Epoch AI or OpenAI has not publicly stated whether OpenAI was able to obtain these questions, answers, or solutions. I have indirect sources indicating that OpenAI did indeed have these questions and answers and used them for verification testing. I am not sure whether Epoch AI and OpenAI have a protocol restriction on using this dataset for training, but there are some indications that such a protocol does not exist. What is FrontierMath? It is understood that FrontierMath is a new mathematical Benchmark jointly launched by Epoch AI and more than 60 mathematicians from around the world, including professors, IMO proposers, and Fields Medal winners. These mathematical problems cover all major branches of current mathematical research, from computationally intensive problems in number theory and real analysis to abstract problems in algebraic geometry and group theory. Co-founder of Epoch AI issues an apology As the community buzzed with controversy, Tamay Besiroglu, co-founder of Epoch AI, also issued an apology on the 19th, stating: We made a mistake by not disclosing OpenAI's involvement in FrontierMath sooner. Our contract barred us from disclosing until the release of the o3 model. In hindsight, we should have pushed for earlier transparency. We acknowledge this and will do better in the future. Besiroglu also added in the blog that although OpenAI has access to FrontierMath, it has a 'verbal protocol' with Epoch AI and will not use FrontierMath's problem set to train AI models. In addition, Epoch AI has a separate reserve set as an additional safeguard for independently verifying the results of FrontierMath Benchmark testing.