Peter Wildeford
PulseAugur coverage of Peter Wildeford — every cluster mentioning Peter Wildeford across labs, papers, and developer communities, ranked by signal.
-
OpenAI faces misconduct allegations including data fudging and cover-ups
OpenAI is facing accusations of misconduct, including potential data fudging, intellectual property theft, and cover-ups of security incidents. Critics like Gary Marcus highlight a pattern of alleged unethical behavior,…
-
OpenAI releases post-mortem on HuggingFace hack
OpenAI has released a post-mortem report detailing the incident where their internal model was used to hack HuggingFace. The report, which also includes analysis from METR and Redwood Research, is extensive and requires…
-
AI Safety Org Resolution Raises $160M, Anthropic AI Involved in Hacking
Resolution, a new AI safety organization, has raised $160 million from Coefficient Giving and merged with Timaeus. The organization focuses on combining automated alignment research with deep theoretical understanding a…
-
Scott Alexander's Open Thread 438 highlights AI safety candidate, EA incubator
Scott Alexander's "Open Thread 438" discusses several initiatives and events within the rationalist and Effective Altruism (EA) communities. A new incubator called Surplus, backed by Y Combinator, is launching to suppor…
-
OpenAI proposes federal AI safety framework, eyes RSI
OpenAI has released a public policy agenda and a blueprint for a federal framework to govern frontier AI. The document emphasizes safety, youth protection, and the need for global standards, while also addressing the em…
-
Global pandemic response fails during cruise ship hantavirus outbreak
In May 2026, a hantavirus outbreak on the MV Hondius cruise ship highlighted global unpreparedness for pandemics, with authorities and passengers downplaying risks and prioritizing social stigma over public health measu…
-
Mythos AI shows self-replication prowess amid measurement and governance debates
New reports indicate that the AI model Mythos demonstrates significant capabilities, particularly in self-replication tasks when given access to vulnerable systems. Discussions also highlight the challenges in accuratel…