Edit Answer: SlimeBunnyBat's Answer to Andy Gee's question on Mesa-Optimizers 2

From Stampy's Wiki
SlimeBunnyBat's Answer to Andy Gee's question on Mesa-Optimizers 2
Log-in is required to edit or create pages.

You do not have permission to edit this page, for the following reason:

The action you have requested is limited to users in the group: Users.

Andy Gee's question on Mesa-Optimizers 2

@3:54 you mention providing the whole of Wikipedia for learning data. Wikipedia details several methods for breaking memory containment. If this is provided to an advanced AI couldn't that AI become aware that it may me constrained within blocks of memory, and thus attempt to bypass those constraints to maximize it's reward function?
These vulnerabilities have been present in all Intel and AMD CPUs for 20+ years before discovery and have been largely mitigated, however the "concept" of looking for vulnerabilities in micro architecture is something an AI can do a lot better than humans can. If you read the Assembly for pre-forking in Intel chips, it's pretty obvious the entire memory space is available while the CPU is predicting what will be required of it next.
Presuming containment of an AI system is important, isn't feeding massive datasets a considerable risk, not only for intellectual property rights but to maintain control of the AI?

Here's some examples of existing vulnerabilities, who knows how many more there are.

AnswerText of the answer.
TagsWhat topics are associated with this answer?
Related questions:Question(s) which this is related to.
Answer to:What question is this an answer to?
Brief Answer:Text of the brief version of the answer. Only shows when the normal answer is over 1500 characters.
AuthorWho is the primary author of this answer?
DateWhen was this answer written?
SourceOnly use this if you copied the answer from an external source, like an FAQ (and be sure to ask permission first!)
Source URL: