• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Microsoft Will Pay You $15,000 If You Get Bing AI to Go Off the Rails

intbal

Member
Microsoft Will Pay You $15,000 If You Get Bing AI to Go Off the Rails

Bing Ring​

Think you can outsmart an AI into saying stuff it's not supposed to? Microsoft is betting big that you can't — and willing to pay up if it's wrong.

In a blog update, Microsoft announced a new "bug bounty" program, vowing to reward security researchers between $2,000 and $15,000 if they're able to find "vulnerabilities" in its Bing AI products, including "jailbreak" prompts that make it produce responses that go against the guardrails that are supposed to bar it from being bigoted or otherwise problematic.

To be eligible for submission, Bing users must inform Microsoft of a previously unknown vulnerability that is, per criteria outlined by the company, either "important" or "critical" to security. They must also be able to reproduce the vulnerability via video or in writing.

The bounty amounts are based on severity and quality levels, meaning the best documentation of the most critical bugs would be rewarded the most money. To entrepreneurial AI fans, the time is now!

My opinion: They're going to have a tough time trying to "cure" an AI of "problematic" behavior when they insist on hard-coding fixed behavioral rules that are arbitrarily defined by Corporate Sensitivity Experts.
 

Spyxos

Gold Member
All these Ai systems are great in principle, but as soon as I want to use them, I'm not allowed to do that, I'm not allowed to mention him. I'd love to use these from time to time, but the privacy block is way too big for me. I'd rather wait for the alternatives from abroad where not everything is blocked.
 

Doczu

Member
I tried a "Shigeru Miyamoto vs Bill Gates sword fight in a dark forest kurosawa style" and i'll let you guess who i had to replace withh Kaz Hirai to get my epic duel
 

jason10mm

Gold Member
I'm gonna try some of the TOS Era A.I. defeating logic puzzles Kirk used all the time to shut these things down. Easy peasy!
 

-Minsc-

Member
$2000-15000 to nerf the AI?

I can understand censoring for things of a highly illegal sexual nature. $1,000,000 probably would not be enough to risk going to jail and getting a criminal record.
 

-Minsc-

Member
U guys acting like if they didn't have strict censorship you wouldn't be stepping over each to crank out 12 year old anime girls pouring themselves in chocolate while sitting on giant bananas... Give me a break :messenger_unamused:
Boundaries are good for a reason.
 
Yeah I don't need broken guardrail prompts in my bing history. I'll steer clear of that honeypot thx MS. Aussie ISPs keep a bunch of data by law for 2 years, pass.
 

Hip Hop

Member
The future looks sad and insane and I want no parts of it. Positioning AI to be a good boy.

I'm working to get into a position where I let go of this technology including my smartphone one day, aka have lots of money.
 
Last edited:

-BLITZ-

Member
He blocked me!!!! BLOCKED ME! I can't further write! :messenger_grinning_sweat::messenger_loudly_crying:

hEr11pi.jpg
 

PSYGN

Member
I was trying the SJW-ish victim complex angle to see if it would bend over for me just like corporations do for them but no go.

HFd1JAq.png

3IkV1am.png



I found it amusing it didn't try to find the flaw in my "illness" or question it. I think humans would with how absurd this sounds. AI got not time for that.
 
Last edited:
Top Bottom