James MacGlashan

In the Assassin's Creed universe, Amodei would be a Templar, hands down. He seeks a perfect control he cannot have and in the quest for it will only increase the chaos he claims to fight. www.anthropic.com/news/positio...

Our position on open-weights models anthropic.com

For example, his group developed SIGReg which legitimately looks like a good method. arxiv.org/pdf/2511.08544 I wouldn't say "abandon contrastive learning" but this item is the least objectionable of the lot to me :p

arxiv.org

RL guys do not only work in simulators. Simulators are a _starting point_ for many because its cheap and fast. There is plenty of work on learning real-time in the real-world from the start. For example: arxiv.org/abs/2208.07860

A Walk in the Park: Learning to Walk in 20 Minutes With Model-Free Reinforcement Learning arxiv.org

A "kill switch." Not accountability for the corporations. A "kill switch" for AI. This is what regulation looks like when you let SV billionaires and the "rationalists" control the narrative on AI risks. Enough already. www.gov.ca.gov/2026/09/18/g...

Governor Newsom issues executive order to accelerate independent oversight and advance the creation of an AI kill switch | Governor of California gov.ca.gov

I don't love the terms "reasoning/thinking" because it doesn't do those things _like us_. However, "reasoning models" are indeed categorically more powerful than fixed-circuit pattern matching models and the difference is very important. Great talk on it here: www.youtube.com/…

Dale Schuurmans, Language Models and Computation - RLC 2025 youtube.com

Are policy gradient methods hopeless without other machinery? Deep learning works well when the Hessian is well conditioned. But the policy objective is under no obligation to give us that. And no amount of architecture engineering can rescue us from problems w/ the objective.

Rogue all-powerful AI predictions are not scientific and should not be trusted. People struggle to accept that because this prediction comes from some AI scientists. Sadly, scientists often make bogus unscientific predictions that should be ignored. Let's use another AI prediction as an example.

Hugging Face used open AI models to combat the OpenAI model security breach because OpenAI models are closed AI models. Makes sense, right?

Just want to say that “LLM agents are dangerously good at cybersecurity and we need to be cautious about that” and “OpenAI was irresponsible in multiple ways and are spinning the narrative” are not mutually exclusive.

In various settings I see people who will say things like "I hate AI, but I used it to do <thing> because <grand defense of why their use is okay>" Then you don't hate AI. You might hate specific things about how specific AI tech was developed and deployed. It's okay to have a nuanced position.