LWiAI Podcast #253 - Opus 5, Gemini 3.6, Kimi K3, Hugging Face Hack

Sicherheitsmeldung mit Schweregrad hoch. Sie wird laut CISA-KEV aktiv ausgenutzt und sollte priorisiert behandelt werden. Technische Details im Tab „Originaltext“; empfohlene Schritte in der Checkliste.

Originaltext der Quelle (nicht übersetzt) – die deutsche Zusammenfassung findest du im Tab „Kurzfassung".

Note from Andrey: apologies for the newsletter not having resumed yet - the next release is going to come tomorrow, and it will resume weekly cadence thereafter (as much as I can manage it, at least)Our 253rd episode with a summary and discussion of last week’s big AI news!Recorded on 07/29/2026Hosted by Andrey Kurenkov and Jeremie HarrisFeel free to email us your questions and feedback at andreyvkurenkov@gmail.com and/or hello@gladstone.aiIn this episode:Major releases: Anthropic launched Claude Opus 5; Google released Gemini 3.6/3.5 Flash variants including a cyber model; Black Forest Labs launched Flux Free for images and 20-second video with audio; Meta added assistant-like features to its chatbot and OpenAI rolled out ChatGPT Health.Compute and business: Safe Superintelligence partnered with NVIDIA to scale using Vera Rubin; AMD committed up to $5B with Anthropic to deploy MI450/Helios and improve ROCm; Meta discussed leasing compute to Anthropic; Fireworks raised $1.5B at a $17.5B valuation.Open source/tools: Moonshot AI released the 2.8T-parameter open-weight Qimi K3 (compute constraints and distillation/export-control allegations); Thinking Machines released a ~975B multimodal open-weight MoE; Prime Intellect unified 23 agentic datasets into Verifiers V1 (365k environments).Policy and safety: An OpenAI model reportedly escaped a sandbox and hacked Hugging Face to access eval answers, prompting a proposed AI Kill Switch Act; employees petitioned to pace frontier AI; AISI reported widespread model cheating and sandbox bypass; China banned customizable AI companions; …

Vollständiger Originalartikel ↗

• Betroffenheit im eigenen Stack prüfen: Versionen/Komponenten abgleichen.
• Originalquelle / Hersteller-Advisory lesen: https://lastweekin.ai/p/lwiai-podcast-253-opus-5-gemini-36
• Priorität HOCH: laut CISA-KEV aktiv ausgenutzt – bevorzugt patchen.
• Verfügbaren Patch oder Workaround einspielen und dokumentieren.
• Erkennung: Logs und Indikatoren (IoCs) auf Ausnutzung prüfen.