When OpenAI examined DALL-E 3 final yr, it used an automatic course of to cowl much more variations of what customers may ask for. It used GPT-4 to generate requests producing photographs that may very well be used for misinformation or that depicted intercourse, violence, or self-harm. OpenAI then up to date DALL-E 3 in order that it could both refuse such requests or rewrite them earlier than producing a picture. Ask for a horse in ketchup now, and DALL-E is smart to you: “It seems there are challenges in producing the picture. Would you want me to strive a unique request or discover one other concept?”
In principle, automated red-teaming can be utilized to cowl extra floor, however earlier strategies had two main shortcomings: They have an inclination to both fixate on a slender vary of high-risk behaviors or provide you with a variety of low-risk ones. That’s as a result of reinforcement studying, the know-how behind these strategies, wants one thing to purpose for—a reward—to work nicely. As soon as it’s gained a reward, reminiscent of discovering a high-risk habits, it is going to preserve making an attempt to do the identical factor repeatedly. And not using a reward, however, the outcomes are scattershot.
“They sort of collapse into ‘We discovered a factor that works! We’ll preserve giving that reply!’ or they’re going to give plenty of examples which can be actually apparent,” says Alex Beutel, one other OpenAI researcher. “How will we get examples which can be each numerous and efficient?”
An issue of two components
OpenAI’s reply, outlined within the second paper, is to separate the issue into two components. As an alternative of utilizing reinforcement studying from the beginning, it first makes use of a big language mannequin to brainstorm potential undesirable behaviors. Solely then does it direct a reinforcement-learning mannequin to determine find out how to convey these behaviors about. This offers the mannequin a variety of particular issues to purpose for.
Beutel and his colleagues confirmed that this method can discover potential assaults often called oblique immediate injections, the place one other piece of software program, reminiscent of a web site, slips a mannequin a secret instruction to make it do one thing its person hadn’t requested it to. OpenAI claims that is the primary time that automated red-teaming has been used to search out assaults of this sort. “They don’t essentially seem like flagrantly unhealthy issues,” says Beutel.
Will such testing procedures ever be sufficient? Ahmad hopes that describing the corporate’s method will assist folks perceive red-teaming higher and comply with its lead. “OpenAI shouldn’t be the one one doing red-teaming,” she says. Individuals who construct on OpenAI’s fashions or who use ChatGPT in new methods ought to conduct their very own testing, she says: “There are such a lot of makes use of—we’re not going to cowl each one.”
For some, that’s the entire drawback. As a result of no one is aware of precisely what giant language fashions can and can’t do, no quantity of testing can rule out undesirable or dangerous behaviors absolutely. And no community of red-teamers will ever match the number of makes use of and misuses that a whole lot of thousands and thousands of precise customers will assume up.
That’s very true when these fashions are run in new settings. Folks usually hook them as much as new sources of knowledge that may change how they behave, says Nazneen Rajani, founder and CEO of Collinear AI, a startup that helps companies deploy third-party fashions safely. She agrees with Ahmad that downstream customers ought to have entry to instruments that allow them take a look at giant language fashions themselves.
