One prediction is that if you continued training one a wide variety of general LLM examples (task 1) during the final stage you wouldn’t get “emergent misalignment”. It’s the task 1-task 2 sequencing that allows forgetting of task 2.
ETHICS
-
Catastrophic Forgetting in Neural Networks and LLM Safety
By
–
“Catastrophic forgetting” is a misnomer, it’s not usually actually catastrophic for neural nets. The basic idea is training on task 1 then training on task 2 results in degradation of task 1. Here task 1 is being a good LLM and task 2 is writing dangerous code.
-
Using Claude to Recreate Famous TV Show Sets
By
–
Using Claude to recreate famous sets from TV shows.
— Allie K. Miller (@alliekmiller) 28 février 2025
I fear I have done something unspeakable. https://t.co/XYCk0vyYRy pic.twitter.com/5v3F6azNRsUsing Claude to recreate famous sets from TV shows. I fear I have done something unspeakable.
-
Defending Em Dashes and Oxford Commas Against AI Detection Myths
By
–
You can pry my —em dash and my oxford comma from my cold, dead, non-AI hands. If you want to tell if AI actually wrote something, these are not the signs to look for. (Nor are they 'rarely used')
-
When does the AI-as-intern analogy fail us?
By
–
We keep saying AI is like a very smart intern: give it more background & context, show it good examples, review its work, assume it has low expertise, assume it won’t be perfect. When is this analogy shooting us in the foot?
-
Thales AI-Powered Ship Identification Keeps Humans Central
By
–
We’re inspired by the innovations our customers & partners create with Metis AIPU! Check out @thalesgroup AI-powered approach to ship identification at sea, keeping humans at the center while enhancing decision-making. #AI #EthicalAI #Innovation
-
Objective AI preference tests lack real world validity context
By
–
This is interesting but it means less than you might assume. “Objective” preference tests like this don’t measure much—context deeply informs preference. Picking one piece of text vs another in a poll is a measure of preference in a peculiar non real world scenario. What
-
Generative AI in Healthcare: Benefits and Challenges
By
–
Generative AI In Healthcare: The Benefits And Challenges
#AI #AIio #BigData #ML #NLU #Futureofwork @HaroldSinnott @fogoros @iainljbrown @NandoDF @katecrawford @drhassanrashidi @YuHelenYu http://
ow.ly/e8PR30sK3la -
GPT-5 and the Loss of Human Decision-Making Autonomy
By
–
🤖 Why GPT-5 Sparks Concern.
— Bernard Marr (@BernardMarr) 28 février 2025
It’s not about job replacement—it’s about decision-making autonomy. As AI takes over more decisions, we risk losing our ability to make complex choices. Learn how to stay in control. 🧠
📌 Key Topics:
* Decision-making in the AI era
* The risks of… pic.twitter.com/Cdeg3dzsXxWhy GPT-5 Sparks Concern. It’s not about job replacement—it’s about decision-making autonomy. As AI takes over more decisions, we risk losing our ability to make complex choices. Learn how to stay in control. Key Topics:
* Decision-making in the AI era
* The risks of -
Open-Source AI Critical for Safety and Humanity Resilience
By
–
Holy crazy fu**ing sh*t! My TED talk is finally out In this short story I explain why open-source AI is critical for safety and humanity's resilience. Yep, you don’t hear that perspective very often! It's called "What if AI just works" Find here: