
Anthropic Says AI Is Speeding Up Its Own Research (and llama.cpp 1.0)
Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! Anthropic, the company behind Claude, told US securities regulators it has met its own threshold for AI speeding up AI research, and an outside group got a rare look inside. Also in this episode: the FBI warns about ARTEX, an AI hacking tool that took some attackers from first look to admin access in under three hours, and why its advice isn't really about AI. And llama.cpp, the free, open-source engine for running AI on your own computer, hits version 1.0. Prices: llama.cpp is free and open source. Sources: SEC: Anthropic Form S-1/A, Amendment No. 2: https://www.sec.gov/Archives/edgar/data/0002130357/000162828026072484/anthropicpbc-sx1ax2.htm Anthropic: Responsible Scaling Policy: https://www.anthropic.com/responsible-scaling-policy METR: Independent Assessment of AI R&D Acceleration at Anthropic: https://metr.org/blog/2026-11-05-anthropic-ai-rd-acceleration-assessment FBI FLASH on ARTEX: https://www.ic3.gov/CSA/2026/261113.pdf TechCrunch: https://techcrunch.com/2026/11/16/fbi-says-artex-ai-open-source-hacking-tool-used-to-breach-networks-in-under-three-hours/ llama.cpp v1.0.0 release notes: https://github.com/ggml-org/llama.cpp/releases/tag/v1.0.0 Watch the video version: https://youtu.be/G5gCgFmgnys Got something wrong? Tell me: hello@bluebraintech.ca Music made with Suno.
Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.
