That's such a silly take, just completely disconnected from objective reality. There's no need for more AI safety research of the type you describe. There researchers who want more money for AI safety are mostly just grifters trying to convince others to give them money in exchange for writing more alarmist tweets.

If systems can be hacked then they will be hacked. Whether the hacking is fine by an AI, a human, a Python script, or monkey banging on a keyboard is entirely irrelevant. Let's focus on securing our systems rather than worrying about spurious AI risks.