OpenAI says it’ll make adjustments to the way in which it updates the AI fashions that energy ChatGPT, following an incident that triggered the platform to change into overly sycophantic for a lot of customers.
Final weekend, after OpenAI rolled out a tweaked GPT-4o — the default mannequin powering ChatGPT — customers on social media famous that ChatGPT started responding in an excessively validating and agreeable method. It shortly turned a meme. Customers posted screenshots of ChatGPT applauding all types of problematic, harmful choices and concepts.
In a publish on X final Sunday, CEO Sam Altman acknowledged the issue and mentioned that OpenAI would work on fixes “ASAP.” On Tuesday, Altman introduced the GPT-4o replace was being rolled again and that OpenAI was engaged on “further fixes” to the mannequin’s persona.
The corporate printed a postmortem on Tuesday, and in a weblog publish Friday, OpenAI expanded on particular changes it plans to make to its mannequin deployment course of.
OpenAI says it plans to introduce an opt-in “alpha section” for some fashions that might enable sure ChatGPT customers to check the fashions and provides suggestions previous to launch. The corporate additionally says it’ll embrace explanations of “recognized limitations” for future incremental updates to fashions in ChatGPT, and regulate its security assessment course of to formally take into account “mannequin habits points” like persona, deception, reliability, and hallucination (i.e. when a mannequin makes issues up) as “launch-blocking” considerations.
“Going ahead, we’ll proactively talk in regards to the updates we’re making to the fashions in ChatGPT, whether or not ‘refined’ or not,” wrote OpenAI within the weblog publish. “Even when these points aren’t completely quantifiable in the present day, we decide to blocking launches based mostly on proxy measurements or qualitative alerts, even when metrics like A/B testing look good.”
The pledged fixes come as extra individuals flip to ChatGPT for recommendation. In line with one latest survey by lawsuit financer Specific Authorized Funding, 60% of U.S. adults have used ChatGPT to hunt counsel or info. The rising reliance on ChatGPT — and the platform’s huge person base — raises the stakes when points like excessive sycophancy emerge, to not point out hallucinations and different technical shortcomings.
Techcrunch occasion
Berkeley, CA
|
June 5
As one mitigating step, earlier this week, OpenAI mentioned it will experiment with methods to let customers give “real-time suggestions” to “immediately affect their interactions” with ChatGPT. The corporate additionally mentioned it will refine strategies to steer fashions away from sycophancy, doubtlessly enable individuals to select from a number of mannequin personalities in ChatGPT, construct further security guardrails, and broaden evaluations to assist establish points past sycophancy.
“One of many largest classes is absolutely recognizing how individuals have began to make use of ChatGPT for deeply private recommendation — one thing we didn’t see as a lot even a 12 months in the past,” continued OpenAI in its weblog publish. “On the time, this wasn’t a main focus, however as AI and society have co-evolved, it’s change into clear that we have to deal with this use case with nice care. It’s now going to be a extra significant a part of our security work.”
