Take AISafety.info’s 3 minute survey to help inform our strategy and priorities

Take the survey

Aren't there easy solutions to AI alignment?

If you've been learning about AI alignment, you might have had a thought like, "Why can't we just do [this thing that seems like it would solve the problem]?"

Unfortunately, many AI alignment proposals turn out to have hidden difficulties. It’s surprisingly easy to come up with “solutions” that don’t actually solve the problem. Some intuitive alignment proposals (that are generally agreed to be flawed, incomplete, or hard to implement) include:

Some common ways alignment proposals fail are that the proposed solution:

See also John Wentworth’s "Why Not Just…" sequence.



AISafety.info

We’re a global team of specialists and volunteers from various backgrounds who want to ensure that the effects of future AI are beneficial rather than catastrophic.