If the AIs in question are moral patients
13 February 2025 (Joe Carlsmith's Substack). Joe Carlsmith in his February 2025 Substack essay "What is it to solve the alignment problem?", on defining the AI alignment problem, discussing whether superintelligent AIs could have moral status.
What is it to solve the alignment problem?
More from Joe Carlsmith · AI and philosophy · AI and safety · 2025
Who else?
What it refers to: The techniques for preventing rogue behavior in AI systems discussed in the series, which would raise disturbing ethical concerns if the AIs are moral patients.
Other voices
- “Not because we know the system is conscious, but because we can't know it isn't” Grigori Guitchounts,
- “If AI has consciousness, then we need to treat it as a moral subject is the wrong way of thinking” Blaise Agüera y Arcas, Researcher,
- “We must not train them to act as though they do” Mustafa Suleyman, Builder,
- “Creating conscious machines would be an ethical disaster” Anil Seth, Researcher,
- “Can a machine have toothache?” Ludwig Wittgenstein,










