| ▲ | rvz 14 hours ago |
| This is why "AI safety" is a complete joke to these companies. |
|
| ▲ | danpalmer 9 hours ago | parent | next [-] |
| All of the "hacks" (from Google and others) were the same issue – the vendor Irregular running the models unsandboxed. |
|
| ▲ | pixl97 12 hours ago | parent | prev | next [-] |
| Because they got complacent only having incompetent AIs. Real life is like reading a fiction story where a group of people are trying to prevent an unfolding disaster. The crazy thing is they already have the instructions on how to prevent the disaster. And they go on to ignore the instructions and with every step make the problem worse. Most people would put it down as being too cliche. |
|
| ▲ | talon8635 12 hours ago | parent | prev | next [-] |
| Well, to be fair, isn’t it an unsolved question? Are they constructing sandboxes, signaling intent to be safe, but their own models are smarter than their internal security team building the sandbox? |
| |
| ▲ | arcfour 11 hours ago | parent [-] | | As a security engineer I have no idea why these sandboxes would even be connected to the internet at all for tasks that aren't intended to use the internet. A package proxy? Why not run our own internal cache? Then we aren't at (as great a) risk of someone poisoning it with a malicious package during model training, for example... | | |
| ▲ | borski 10 hours ago | parent | next [-] | | We’re hiring. :) (And we’re fixing many of these things, but worth noting this happened at a third party vendor, not in our lab) | | |
| ▲ | xnx 8 hours ago | parent [-] | | Could you add any detail on why Google uses (used?) Irregular? I wouldve thought that type of service would be a core competency that Google needs internally. | | |
| ▲ | borski 6 hours ago | parent [-] | | Even if you had it internally (which we do), there is so much surface area and it’s such a novel space that you’d want as much testing on it as possible. There aren’t many vendors, and irregular is one. |
|
| |
| ▲ | eli_gottlieb 7 hours ago | parent | prev [-] | | AFAICT one fundamental issue is that they don't seem to have hired actually security engineers or experts to do any actual security. |
|
|
|
| ▲ | verdverm 13 hours ago | parent | prev [-] |
| they're all using the same vendor, re: incestuous EA cabal |
