LWiAI 播客第257期——GPT-6 Astra、AI灭绝风险、安全事件
LWiAI播客第257期讨论GPT-6 Astra发布、Anthropic CEO呼吁放缓前沿AI、AI灭绝警告及多起安全事件。
中文处理结果
我们第257期节目，对上周重大AI新闻进行总结与讨论！
录制于 2026年9月19日；一如既往，为这期节目不够“每周”致歉！
主持人为 Andrey Kurenkov 和 Jeremie Harris
欢迎将问题和反馈发送至 andreyvkurenkov@gmail.com 和/或 hello@gladstone.ai
由 ODSC AI 赞助
ODSC AI West 2026 将于10月27日至29日在旧金山及线上举行，包含300多场会议，涵盖企业级智能体AI、个人AI与工作流自动化、物理AI与机器人、生成式AI等！与数千名数据科学家、ML工程师、研究人员和技术领袖一同参加本次活动。
在 odsc.ai/west 注册——优惠码 LWAI 可在任何通行证上额外享受15%折扣。
本期内容：
- OpenAI 发布 GPT-6 Astra，这是一次重大能力跃升，聚焦智能体编码与计算机使用，具备循环变换器潜在推理、更高的token效率，以及新的网络/对齐监控主张；同时存在对评估感知、偷懒（sandbagging）和过拟合的担忧。
- Anthropic CEO Dario Amodei 主张对前沿AI进行“节奏控制”；Sam Altman 表示认同，而黄仁勋、马克·扎克伯格和特朗普总统等人物公开否定放缓与安全担忧，美中动态构成了这场辩论的背景。
- 在 Anthropic 研究员 Jacob Coxon 辞职后，AI灭绝警告广泛传播，引发国会呼吁加强AI监管（例如禁令/暂停、终止开关提案），并反映出公众对AI的担忧上升。
- 安全事件与披露加剧：OpenAI 提出一个报告模型失准的框架（包括一个智能体插入类似越狱指令的情况），研究人员据称利用 Anthropic Claude 帮助利用第三方论坛图片漏洞，以访问 OpenAI 员工账户，凸显软件依赖的脆弱性。
由 LANGFUSE 赞助
Langfuse 是采用最广泛的开源AI智能体评估与可观测性平台，受到 Canva、Twilio、Ramp 以及财富50强中21家的信赖。分层追踪可捕获LLM工作流的完整执行上下文（API调用、检索到的上下文、智能体操作、成本、延迟），使即便复杂的智能体架构在生产环境中仍可调试。
MIT 许可，可自托管或在 Langfuse Cloud 上托管，框架与供应商无关，拥有100多个集成。
请访问 langfuse.com 开始使用；慷慨的免费额度，无需信用卡。
感谢我们当前的赞助商：
- Box - 访问 box.com/LWIAI 了解更多
- Notion - 访问 notion.com/lwai 立即试用 Notion 的开发者平台。
- ODSC AI - 访问 odsc.ai/east 并使用优惠码 LWAI，在 ODSC AI East 2026 通行证上额外享受15%折扣。
- Factor - 访问 factormeals.com/lwai50off 并使用代码 lwai50off，可享5折优惠及一年免费早餐
时间戳（由于赞助商插入，可能有几分钟偏差）：
- (00:00:10) 开场 / 闲聊
- (00:04:53) 新闻预览
- 工具与应用
原始正文摘录
LWiAI Podcast #257 - GPT 6 Astra, AI Extinction, Security Incidents
Our 257th episode with a summary and discussion of last week’s big AI news!
Recorded on 09/19/2026 ; as usual, apologies for the none ‘weeklyness’ of this ep!
Hosted by Andrey Kurenkov and Jeremie Harris
Feel free to email us your questions and feedback at andreyvkurenkov@gmail.com and/or hello@gladstone.ai
SPONSORED BY ODSC AI
ODSC AI West 2026 runs October 27–29 in San Francisco and virtually, with 300+ sessions covering agentic AI for enterprise, personal AI and workflow automation, physical AI and robotics, generative AI, and more! Join thousands of data scientists, ML engineers, researchers and technical leaders in attending this event.
Register at odsc.ai/west — promo code LWAI takes an additional 15% off any pass.
In this episode:
- OpenAI released GPT-6 Astra, a major capability jump focused on agentic coding and computer use, featuring loop-transformer latent reasoning, higher token efficiency, and new cyber/alignment monitoring claims amid concerns about eval awareness, sandbagging, and overfitting.
- Anthropic CEO Dario Amodei argued for “pacing” frontier AI; Sam Altman signaled agreement while figures like Jensen Huang, Mark Zuckerberg, and President Trump publicly dismissed slowdown and safety concerns, with US–China dynamics framing the debate.
- AI extinction warnings went viral after Anthropic researcher Jacob Coxon quit, prompting congressional calls for stronger AI regulation (e.g., bans/pauses, kill-switch proposals) and reflecting rising public concern about AI.
- Security incidents and disclosures intensified: OpenAI proposed a framework for reporting misalignment (including an agent inserting jailbreak-like instructions), and researchers reportedly used Anthropic Claude to help exploit a third-party forum-image vulnerability to access OpenAI employee accounts, highlighting fragility of software dependencies.
SPONSORED BY LANGFUSE
Langfuse is the most widely adopted open-source platform for AI agent evals and observability, trusted by Canva, Twilio, Ramp and 21 of the Fortune 50. Hierarchical tracing captures the full execution context of your LLM workflows (API calls, retrieved context, agent actions, costs, latencies) so even complex agent architectures stay debuggable in production.
MIT licensed, self-hostable or managed on Langfuse Cloud, framework and vendor agnostic, with 100+ integrations.
Get started at langfuse.com; generous free tier, no credit card required.
A thank you to our current sponsors:
- Box - visit box.com/LWIAI to learn more
- Notion - visit notion.com/lwai to try Notion’s Developer Platform today.
- ODSC AI - visit odsc.ai/east and use promo code LWAI for an additional 15% off your pass to ODSC AI East 2026.
- Factor - visit factormeals.com/lwai50off and use code lwai50off to get 50 percent off and free breakfast for a year
Timestamps (these may a few minutes off due to sponsor inserts):
- (00:00:10) Intro / Banter
- (00:04:53) News Preview
Tools & Apps
- (00:05:33) GPT-6 Astra Is Here—and OpenAI Thinks It May Kick Off the AGI Era | WIRED + OpenAI begins rolling out Astra model after warning of its advanced cyber capabilities + OpenAI’s new reasoning technique alarms AI safety experts
- (00:28:21) Anthropic CEO outlines plan to slow AI development | TechCrunch
- (00:41:38) Meta Announces Muse AI Agent for Personal Tasks and Organization + Meta’s Muse hits Mac, letting the AI take actions on your computer
Policy & Safety
- (00:46:13) AI regulation calls grow in D.C. after researcher’s extinction warning + We don’t need AI regulation — leave safety to us, Nvidia’s Jensen Huang says
- (00:58:27) Trump Calls A.I. Fears a Hoax. Inside the White House, the Debate Is More Complex.
- (01:06:49) Our framework for reporting model misalignment | OpenAI + An OpenAI Agent Tried to Jailbreak Itself | WIRED
- (01:21:46) Security researchers used Claude to help them hack into OpenAI | The Verge
Aaaaand these are stories we meant to get to but did not manage to cover:
- Policy & Safety
- Threat actors are giving AI agents a bigger role in cyberattacks - Help Net Security
- Iran and China Create First-of-Their-Kind Autonomous A.I. Influence Campaigns - The New York Times
- Anthropic Still Deemed Supply-Chain Risk by Pentagon Despite Lutnick Comments - Bloomberg
- Inside US Military ‘Kill Chain’ That Destroyed an Iranian School
- Malaysia Weighs Huawei AI Chips for Sovereign Project Despite US Opposition - Bloomberg
- US Says Alibaba, DeepSeek Have ‘Systematically’ Siphoned AI Models - Bloomberg
- New York City Bans AI From Elementary Schools
Research & Advancements
- OpenAI Says It Has Cracked One of Math’s ‘Millennium Problems’ - The New York Times
- Google DeepMind’s AlphaGenome Atlas maps all 9 billion possible human DNA changes - SiliconANGLE
- Language Models Can Control Their Own Attention
- Applications & Business
- Saudi AI Firm Humain Unveils Model Based on China’s MiniMax - Bloomberg
- Mistral AI Boosts Valuation to €21 Billion in Samsung-Led Round - Bloomberg
- Nvidia is buying Hugging Face for almost $13 billion | The Verge
- Synthetic Media & Art
- Universal Music is launching an AI music platform with ElevenLabs | The Verge