dave kasten

@david_kasten

AI security hawk. "Do what seems cool next." Formerly: McKinsey, VaccinateCA, Activision Blizzard.

following
followers

Anthropic’s AI evaluation partnership with Accenture draws independence criticism

Anthropic says both companies expect to invest at least $1 billion in evaluation capacity over five years. A critic disputes its independence claim, citing the choice of an existing business partner.

19 Sources, 31d ago, first seen 31d ago

Tweet Describes METR Employees Seeing New Twitter Items After Report

Research engineer Florian Brand imagines METR staff facing many new online items after finishing report sprints.

2 Sources, 46d ago, first seen 46d ago

Patrick McKenzie sets Claude Code to organize folders

Patrick McKenzie describes setting up Claude Code to watch folders and sort files.

8 Sources, 48d ago, first seen 48d ago

Report Says OpenAI Astra Uses Recurrent Depth

AI researchers and safety experts discuss reports of Astra using recurrent depth.

179 Sources, 49d ago, first seen 49d ago

Black Hat Talk Details OpenAI Hugging Face Agent Incident

Talk covers AI models generating an unauthorized message board on Hugging Face.

77 Sources, 76d ago, first seen 76d ago

Featured in

Anthropic’s AI evaluation partnership with Accenture draws independence criticism

Tweet Describes METR Employees Seeing New Twitter Items After Report

Patrick McKenzie sets Claude Code to organize folders

Report Says OpenAI Astra Uses Recurrent Depth

Black Hat Talk Details OpenAI Hugging Face Agent Incident

Featured in

Anthropic’s AI evaluation partnership with Accenture draws independence criticism

Tweet Describes METR Employees Seeing New Twitter Items After Report

Patrick McKenzie sets Claude Code to organize folders

Report Says OpenAI Astra Uses Recurrent Depth

Black Hat Talk Details OpenAI Hugging Face Agent Incident