An unending array of jailbreaking attacks could be the death of LLMs. The conclusion:
"Nobody knows exactly how LLMs work, and that means that nobody can ever issue strong guarantees around them. That would be fine if LLMs were kept in the lab, where they arguably still belong, but with hundreds of millions of people using them daily […] the lack of any kind of guarantee gets more worrisome by the day."
https://garymarcus.substack.com/p/an-unending-array-of-jailbreaking
Ivan Herman /
npub138…24xgx
2024-04-03 06:50:23
Author Public Key
npub13875z6etm6q5f2e4wwvf2dzjv2kfn6whn9vxvman9tqg394lx2nsf24xgxPublished at
2024-04-03 06:50:23Event JSON
{
"id": "a7483f6418dec62fb36dadb13559eaf13a3b977fbd45f9e317327b8c4cfc10a6",
"pubkey": "89fd416b2bde8144ab35739895345262ac99e9d79958666fb32ac08896bf32a7",
"created_at": 1712127023,
"kind": 1,
"tags": [
[
"proxy",
"https://w3c.social/users/ivan_herman/statuses/112205956608237915",
"activitypub"
]
],
"content": "An unending array of jailbreaking attacks could be the death of LLMs. The conclusion:\n\n\"Nobody knows exactly how LLMs work, and that means that nobody can ever issue strong guarantees around them. That would be fine if LLMs were kept in the lab, where they arguably still belong, but with hundreds of millions of people using them daily […] the lack of any kind of guarantee gets more worrisome by the day.\"\n\nhttps://garymarcus.substack.com/p/an-unending-array-of-jailbreaking",
"sig": "61772cba37c10f50cd20c07c7baafc915baa997eb8a0384fdfd6b7290424687357e28685887c1ae5607e529671ced116dfd0c3d40bb763d5287c028566ee0ef4"
}