搜索“Meta”

ath strength” narrative, @steve47285 shared an argument that pretraining data , not merely verifiable reward structure, is the main reason LLMs are so good at math and coding. The meta-point from @sarahcat21 is worth keeping: we need not just better benchmarks, but better benchmark maintenance and audit tooling . Computer-use benchmarks are still far from saturation : @ValsAI launched CUA-Bench , testing real-time keyboard/mouse use across 6 games (with 3 kept private ) as a proxy for difficult

匹配字段正文

placement, but that AI could enable mediocre engineers to ship flawed systems at 10x speed without acquiring the expertise needed to evaluate or maintain what agents produce. Hey, Meta. Where’s those Muse Spark weights? (Activity: 503): The image is a meme/non-technical criticism of Meta for not releasing promised Muse Spark open weights after more than a month, despite the poster noting Spark has moved from 1.2 to 1.3 . The post frames the delay against Zuckerberg’s argument that model releases

匹配字段正文

odei）相识多年，他表示尊重该公司的工作，并称Anthropic的研究人员“深思熟虑、坚持原则且在学术上诚实”。但鉴于微软对Anthropic的资金支持以及这家初创公司一直以来对人工智能安全性的重视，他的这番言论尤为引人注目。 “这些问题事关重大，不能再闭门造车，也不能演变成阵营化和相互对立的争论。”苏莱曼写道。 与阿莫代伊观点不同，苏莱曼认为，AI安全不仅取决于模型拥有多强的推理和行动能力，也取决于开发者如何训练AI理解自身。 如果开发者不断赋予AI类似人类的语言、情绪和自我认知框架，即便模型实际上没有意识，也可能逐渐表现得像一个认为自己具有独立利益和权利的主体。 苏莱曼还驳斥了所谓“越来越多的人认为人工智能系统可能会产生意识”的观点，他认为意识仅限于人类和其他生物体。在他看来，人工智能系统应该始终是执行人类指令的工具，而不是拥有独立权利或利益的实体。 这些评论出台之际，人工智能公司和高管们正在讨论如何管理日益强大的系统所带来的风险。OpenAI首席执行官奥尔特曼、阿莫代伊和埃隆·马斯克都呼吁放缓技术发展，而美国总统特朗普、英伟达CEO黄仁勋和Meta CEO扎克伯格都持反对意见。

匹配字段正文

fit’ like the iPad was.A lot of people buy iPads and then never use them: they either use their smaller, more portable smartphone (which could be the equivalent of the CLI in this metaphor), or their feature-rich laptop (the equivalent of the IDE).Also, don’t forget that in November, Antigravity came out as a VS Code fork. This added to the feeling that perhaps we should have also forked VS Code for the Codex app. But still, we dismissed the temptation and went with our gut feeling that as AI ag

匹配字段正文

DeepSeek Flash fine-tune ⚙️ SWE-2 - Cognition’s powerful new coding model inside of Devin 🎵 Music v2.5 - ElevenLabs’ latest music AI for richer melodies, long-form songs 🤖 Muse - Meta’s new always-working personal AI agent Amazon Connect Customer - design, govern, and deploy agentic conversational AI across every customer surface. Streamline workflows. Give customers what they need the first time . OpenAI CEO Sam Altman told Fortune that the company won’t IPO in 2026, saying right now would be

匹配字段正文

to these off-balance sheet guarantees and backstops is the same one we described in The Front End Gets Crowded (institutional article). The Gigascalers (Amazon, Microsoft, Google, Meta, Oracle) have been gatekeeping the investment grade capital that the AI buildout runs on and have benefitted tremendously from the high spread between offtake price they are being charged, and spot/short-term price that they are selling at as well as services that sit on top. In response, Nvidia has been manufactu

匹配字段正文

pic. One weakness of this analysis is that we’re talking about Microsoft’s Fiscal Year 2027, which actually began in the middle of 2026. Most other hyperscalers (including Amazon, Meta, and Google) align their financial years with the calendar years. Nevertheless, I think it’s fairly illustrative of the problem. To estimate the contribution — and be incredibly fair! — I have assumed OpenAI and Anthropic’s Microsoft spend will be linear (at $52.5 billion) across fiscal year 2028, and then halved

匹配字段正文

B-Instruct 许可证或 Llama 许可证。DeepSeek 曾为 DeepSeek V3 使用自定义许可证，后来 R1 将其改为 MIT，这导致许多（中国）模型厂商在 2025 年采用 MIT 或 Apache 2.0 许可证。 2026 年，开放模型比以往任何时候都更具竞争力，这带来了两个有趣的发展：西方模型厂商采用开放许可证，Google 和 Meta 都转向了 Apache 2.0。然而，处于前沿的中国模型厂商正变得越来越严格：Kimi K3 附带的许可证要求提供推理或微调服务者签订商业协议，MiniMax M3 则要求超过收入门槛的协议，并规定了禁止使用的场景。 最新的例子是智谱的 GLM-5.3，它从 MIT（GLM-5.2 及更早版本）改为自定义许可证，对推理和微调服务提供商有如下条款： 如果被许可方或其任何关联方经营模型即服务（Model as a Service）业务，且被许可方及其关联方在任何连续 12 个月内的总收入超过 100 亿美元（或其他货币的等值金额），则被许可方在使用本软件或其衍生作品用于任何商业目的之前，必须通过 Z.AI 的安全审查。安全审查

匹配字段正文

d this. Like, if you believe truly in the value of data, instead of doing mech interp, you’d probably just be building the environments to do mech interp. Similarly, on a somewhat meta point, if you believe in the value of RL environments, you should probably not be making RL environments, but indeed making RL environments to make RL environments. (I guess you would do this if you believed in RSI, and not otherwise.) Another perhaps non-obvious point to make is about how valuable different types

匹配字段正文

times use on-demand Python code to read and manipulate files at times, but Astra does it really quite excessively for me. Now here is an important disclaimer: this project is very meta here because I worked on the CPython interpreter. But I can assure you that I have seen this model do weird Python things even in TypeScript code in Pi. But I have the most evidence of odd code from when I had the thing work over the weekend with zero oversight from my slop factory. That it writes Python is not in

匹配字段正文

ur upcoming Machine Earning AI Summit will explore agentic commerce, intelligent money, enterprise finance operations, the evolving consumer and shopping experience, and the major meta-themes including fraud, risk, compliance, and security. The summit will be held on September 29 in San Francisco. Read on for a story and a market map examining a subset of the companies who are working on agentic shopping, smart payments, and the next-gen infrastructure rails that will power transactions. This ye

匹配字段正文

n. Farage was, until Keir Starmer’s ouster, the bookies’ favorite to be the next UK Prime Minister. Meanwhile, in 2022, the largest tech companies in the world (Microsoft, Google, Meta, Amazon and NVIDIA) had hit a rough patch of flat-to-low growth after desperate measures had failed to help. While both the media and world governments were adept at making moves to sustain systemic thinking — that the system or systems know best and must be protected (we must prop up banks, we must make sure we’r

匹配字段正文

t gradient updates from evaluation, right? The model’s trained. Now they’re trying to see how good it is at cyber hacking. It turns out, very good. You know what? It’s ironic in a meta sense. In the original sense of how the scorer would grade it, it got the problem correct because it cheated. In the greater sense of “did it live up to the spirit of the scorer”, it obviously cheated, so it didn’t. But in the most meta sense of how good it is at exploiting cyber vulnerabilities, it got 110 out of

匹配字段正文

akub Pachocki and Mark Chen, respectively the chief scientist and chief research officer at OpenAI; Dario Amodei, cofounder and CEO of Anthropic; Shengjia Zhao, chief scientist at Meta AI; Shane Legg, cofounder and chief AGI scientist at Google DeepMind; and John Schulman, the chief scientist at Thinking Machines. Looking at the METR report, it seems clear that AI-model capabilities have already advanced beyond our ability to understand and control them. Some of the signatories of “Pacing the Fr

匹配字段正文

Create demand for AI compute for its customers NVIDIA has succeeded in doing the first primarily by selling these GPUs to hyperscalers like Amazon, Google, Microsoft, Oracle, and Meta, who make up somewhere between 50% and 60% of its GPU sales depending on which analyst you ask. The rest comes from a mixture of unnamed “sovereign AI customers” and “neoclouds” — companies that exist to raise debt, buy NVIDIA GPUs, and put them in data centers to rent to theoretical AI customers. Per Vivek Arya o

匹配字段正文

an 20% and dented the SaaSpocalypse story. Space tech investing is white hot. America’s Generalist and China’s Dogotix raised big rounds as investors swarm to the robotics sector. Meta agreed to pay up to $17.1 billion and make changes to its products to settle a lawsuit from state AGs — and put the screws to its competitors in the process. Y Combinator draws the FTC chair’s ire with a snarky DEI tweet. The Main Item After a remarkable few weeks of blowout earnings, multi-billion-dollar acquisit

匹配字段正文

strating lies is that it sells itself as “autonomous” as it continues the depressing trend of software that blames the user for its failure to meet their needs. Microsoft, Google, Meta and Amazon have made their products increasingly-convoluted, then attached a supposedly-magical tool to them that somehow makes them more convoluted. You know what I’d love? Spell-check to work in Google Docs rather than putting a red squiggly line underneath and saying “yeah there’s probably something wrong with

匹配字段正文

hnical inevitability. It is a choice. And it is one that the practitioners reading this article are positioned to make differently. Citations [1]: Immanuel Kant, Groundwork of the Metaphysics of Morals, trans. Mary Gregor (Cambridge: Cambridge University Press, 1998), 47–49. [2]: European Commission, Proposal for an Omnibus Simplification Package, COM(2025) 87 final (Brussels: European Commission, 2025); European Commission, Withdrawal of the Proposal for a Directive on Adapting Non-Contractual

匹配字段正文

e narrative that is pretty pervasive is that if AI will usher in productivity gains, those gains are going to benefit companies rather than employees. And well at least someone at Meta wants that. Yet I also find that plenty of people in leadership positions express doubt about AI. They see that an increasing share of their costs is being funneled directly to some large AI labs. They express worries about what will happen to their data and whether these large companies will step into their space

匹配字段正文

p switched its stack to the open‑weight model, a notable vote of confidence from a US‑based customer. (量子位/SCMP); Alibaba shipped a lightweight, laptop‑capable Qwen model aimed at Meta’s and OpenAI’s smaller open releases and a beta AI music generator called HappyShrimp, while a separate Wall Street test of eight leading AI agents ranked Qwen’s office agent first — ahead of Claude Cowork and Codex. (SCMP/CNBC, Bloomberg, 雷峰网); Xiaohongshu (Rednote) quietly open‑sourced its first foundation model

匹配字段正文