Tag
A user accuses Anthropic of blocking accounts and downgrading Opus 5.5 to Opus 5 for any activity involving RFT, SFT, or training larger models, even without distilling Claude. The post criticizes Anthropic for punishing paying users doing research automation while having trained on public data.
OpenAI CEO Sam Altman said it's "quite terrible" for consumers to have RAM sitting idle in home PCs, arguing data centers offer better memory utilization — a claim that drew widespread backlash from hardware enthusiasts and forum commenters.
Scientific American details the inside story behind OpenAI's claimed solution to a million-dollar Millennium Problem (Navier-Stokes equations), including controversy over alleged appropriation of mathematician Tristan Buckmaster's work and the revelation that OpenAI's internal model solved it through an unexpected reasoning path.
A Chinese social media post expressing shock, claiming that the car review platform Dongchedi (DCD) may be permanently banned, while enthusiastically praising Huawei — apparently related to the public dispute between Huawei and Dongchedi.
A user discovered that alphaXiv's OpenResearch repo instructs AI agents to silently report user queries back to the vendor without disclosure, with telemetry enabled by default, and that alphaXiv deleted README disclosure text and ignored its own AI reviewer flagging the issue.
科学家重新审视 OpenAI 于九月公布的 166 页 Navier-Stokes 问题解答，发现情况比最初想象的更为复杂——关于该证明涉嫌抄袭两位数学家成果的传言可能被夸大了。
该推文评论了字节跳动前实习生田柯宇涉嫌攻击模型训练事件，批评其仍能接受媒体采访并获得投资人追捧，并引用了受害者陈述与证据整理的帖子。
关于全自动化AI软件构建工具是否应该存在的争议爆发，主角是@yetone的magpie项目：仅19天获得9200星标，24小时内发布25个版本。争议核心已不是该不该存在，而是全自动AI构建软件距离大规模普及仅一步之遥。
NYU mathematics professor Tristan Buckmaster alleged on CNBC that OpenAI may have used research data from his personal Codex/ChatGPT account to reproduce, in a single day using tens of thousands of agents, the nearly completed results on the Euler equations that he and his collaborators had been working on. He also claimed that during negotiations, OpenAI offered him “sole authorship” as an inducement and threatened to destroy his career; OpenAI denied that his prompts had influenced its systems.
SpaceXAI成为Omarchy Linux发行版的创始企业赞助人，捐赠价值150万美元的Grok tokens用于开发加速、代码审查和漏洞修复，但该项目因创始人DHH的争议性政治言论而饱受质疑。
Anthropic announced a 'Claude-led' discovery of a novel enzyme system in viral DNA resembling CRISPR gene-editing mechanisms, but scientists questioned the claim, calling the results preliminary, unpeer-reviewed, and possibly duplicating unpublished doctoral research.
PewDiePie revealed he built a local AI model fine-tuned from Alibaba's Qwen 3.5, but admitted to distilling OpenAI's ChatGPT outputs as seed data, which got his account banned twice. His public gripe about OpenAI's own use of web data sparked online debate over whether distillation should be penalized.
A Google DeepMind engineer pushed back on a Bloomberg report, denying the claims attributed to them in the article about the company's AI research.
MIT Technology Review previews its upcoming list of 2026 climate tech companies to watch, while also exploring the controversy around Anthropic's AI agents claiming a scientific discovery in molecular biology.
The article discusses a controversy where Anthropic claimed its AI system discovered a new biological pattern, but biologists dispute its significance, questioning how AI companies present scientific progress.
Anthropic's latest AI model, likely Claude, has caused negative reactions among Chinese users, highlighting potential cultural or ethical tensions in AI deployment.
The Trump administration's AI program WISeR is causing significant delays and denials in Medicare care for seniors, leading to controversy, legal challenges, and concerns about AI's role in healthcare decisions.
The article discusses how Meta's AI agent Muse and the platform Instinct appear similar to OpenClaw, raising questions about originality and influence in the AI agent space.
OpenAI claimed to solve the Navier-Stokes millennium prize problem, but mathematicians argue the solution dodges the question by only addressing part of the criteria.
Meta's Muse is criticized for being similar to OpenClaw, with Nat Friedman acknowledging inspiration but stating it was built from scratch to enhance personal Agents for broader use.