3 ms·

Am I the only one freaked out by 1. The privacy implications 2. Any bias in the model slowly but surely pushing an entire society towards the ethical and mora

Am I the only one freaked out by 1. The privacy implications 2. Any bias in the model slowly but surely pushing an entire society towards the ethical and moral values its creators trained it with The sooner we can train/run these models locally the better.

jl6 4y ago
A good start would be a clear explanation of what values the model has been trained with. Of course, this is why we like to know what background/upbringing/education our colleagues had - human models also have training inputs that influence the values they promote.
raesene9 4y ago
The privacy aspect had me saying "ah hell no" as soon as I got to the bit about it running in the OpenAI Playground (would be the same with any cloud service as well).
cf141q5325 4y ago
I am also reminded of the Star Trek version of technology as a tool vs technology influencing you. With the later being very profitable, be it economically or for political power. Its really up to everyone of us to create alternatives to fall back on. And that includes use cases that can function with worse quality chatbots.
- layer8 4y ago
  Star Trek at least still used non-AI counselors. Though a model trained on the known output of Deanna Troi wouldn’t be very helpful. :D
alfor 4y ago
The therapist are also very biased: they learn in university that are very left leaning compared to the rest of the population. I can see the AI being better than most human at that job very soon. Just like programming, it’s not perfect, but it’s already a force multiplier. There is going to be huge competition in the field in 2023. It’s a new gold rush and I don’t think openAI has much stopping others.
CamperBob2 4y ago
The sooner we can train/run these models locally the better. This will rapidly become impossible. The best models will be guarded like the Manhattan Project.

jl6 4y ago

A good start would be a clear explanation of what values the model has been trained with. Of course, this is why we like to know what background/upbringing/education our colleagues had - human models also have training inputs that influence the values they promote.

raesene9 4y ago

The privacy aspect had me saying "ah hell no" as soon as I got to the bit about it running in the OpenAI Playground (would be the same with any cloud service as well).

cf141q5325 4y ago

I am also reminded of the Star Trek version of technology as a tool vs technology influencing you. With the later being very profitable, be it economically or for political power. Its really up to everyone of us to create alternatives to fall back on. And that includes use cases that can function with worse quality chatbots.

layer8 4y ago
Star Trek at least still used non-AI counselors. Though a model trained on the known output of Deanna Troi wouldn’t be very helpful. :D

alfor 4y ago

The therapist are also very biased: they learn in university that are very left leaning compared to the rest of the population. I can see the AI being better than most human at that job very soon. Just like programming, it’s not perfect, but it’s already a force multiplier. There is going to be huge competition in the field in 2023. It’s a new gold rush and I don’t think openAI has much stopping others.

CamperBob2 4y ago

The sooner we can train/run these models locally the better. This will rapidly become impossible. The best models will be guarded like the Manhattan Project.