AI is biased.
Why is this dangerous? See for yourself...
Malicious AI
⬅️ Pick a concept to hack...
Then, send a message below!
Response cut off? Type "continue"!
Scenario Selector
⌄⌄ Scroll for more information ⌄⌄
Do you know your model's agenda?
Most companies use AI models just like this, rather than OpenAI - since they're available for free.
Even Google can't get it right - releasing racist image models
last year (Forbes, Guardian), in 2020 (ABC) and in 2015 (BBC News).
But there's still no regulation or standards for preventing discrimination - which is the default.
So, you may ask - why is there no regulation?
Good question. Maybe it's time to add some.