Erich Grunewald argues that statements that large language models are mere "stochastic parrots" (and the like) make unwarranged implicit claims about their internal structure and future capabilities. erichgrunewald.com/posts/against-…

Katja Grace (Katja Grace 🔍) finds that the proportion of respondents to her survey of machine learning researchers who believe extremely bad outcomes from AGI are at least 50% likely has increased from 3% in the 2016 survey to 9% in the 2022 survey. aiimpacts.org/how-bad-a-futu…

Open Philanthropy (Coefficient Giving) has announced a contest to identify novel considerations with the potential to influence their views on AI timelines and AI risk. A total of $225,000 in prize money will be distributed across the six winning entries. openphilanthropy.org/open-philanthr…

The Global Priorities Institute (Global Priorities Institute) has released Hayden Wilkinson's (Hayden Wilkinson) presentation on global priorities research. (The talk was given in mid-September last year but remained unlisted until now.) globalprioritiesinstitute.org/hayden-wilkins…

Daniel Filan (Daniel Filan) interviews John Halstead about why he believes climate change doesn't pose an existential risk. podcasts.google.com/feed/aHR0cHM6L…

Luisa Rodriguez and Rob Wilbin from the 80,000 Hours Podcast interview Rob Long (Robert Long) on why large language models probably aren't conscious. 80000hours.org/podcast/episod…

OpenAI (OpenAI) just announced the launch of GPT-4, "a large multimodal model, with our best-ever results on capabilities and alignment." openai.com/product/gpt-4