Japan Today

ChatGPT chief says AI should be regulated by a U.S. or global agency


The requested article has expired, and is no longer available. Any related articles, and user comments are shown below.

© Copyright 2023 The Associated Press. All rights reserved. This material may not be published, broadcast, rewritten or redistributed without permission.

©2024 GPlusMedia Inc.

Login to comment

Pressed on his own worst fear about AI, Altman mostly avoided specifics, except to say that the industry could cause “significant harm to the world” and that “if this technology goes wrong, it can go quite wrong

Altman knows the NSA is in possession of vast stores of data resources trawled from mass surveillance. It could haven't been, or could be used for LLM training.

The capabilities are known to an extent but now that it is in the wild

will the socioeconomic impact be dealt with in the least harmful way is the question.

0 ( +0 / -0 )

Altman proposed the formation of a U.S. or global agency that would license the most powerful AI systems and have the authority to “take that license away and ensure compliance with safety standards.”

A rare instance when a business chief, Democrats, and Republicans in Congress all agree with each other that regulation is needed

2 ( +2 / -0 )

“Regulated”= programmed by humans to give “correct” responses.

1 ( +2 / -1 )

It all depends. For Trump's TrumpBook social media platform they could program to give it "correct" answers like there is no such thing as environmental damage by human activity and that Biden stole the election....as well as the real truth about PizzaGate, Benghazi, and Hillary's 300,000 emails. In that case call it the TrumpGPT version.

0 ( +0 / -0 )

Regulations contradict to the purpose. AI results based on regulation, preset biases, constraints or shifted weights are just useless. Instead, I would suggest that all results have to be extended by an including digital watermark, for example a blockchain created sequence that identifies the AI origin of any such results itself , the creator and user and the kind of AI system used , timestamp, location of processing and all such. That would allow an AI to run unrestricted as it is meant to be, but still being able to tell if a result is from an AI to prevent misuse or fakes of generated text, sound, pictures, video etc.

1 ( +1 / -0 )

Some people claim that 1 human is infallible while the rest of us are fallible. If AI is truly going to mimic humans it needs to be fallible and every once in a while say something that may offend somebody.

I wonder what a legal AI would have to say about abortion. Would it adopt the moral principals of a protestant, a fundamentalist or Sikhism? Would a medical AI bot advise a woman to get an abortion in NYC but advise against getting an abortion to a woman in Alabama?

Does your phone talk to you in Standard Japanese or Osaka-ben?

0 ( +1 / -1 )

I fear that its already to late, thats why elan musk , jeffrey hinton, sam altman are all suddenly appearing warning of the dangers , when they themselves have invested in it and produced it !!!! the cat is probably already out of the bag, the horse has already bolted....Human kind....Just shut the gate behind you......Bye !

0 ( +0 / -0 )

I wouldn't trust anyone in the US government or a global corporate giant to oversee AI...

0 ( +1 / -1 )

Current and future AI can only be regulated by further AI, so thus Pandora's box is open.

The rate of change from here on in is going to be even more dizzying.

0 ( +0 / -0 )

God don't let the Americans control it... needs to be something along the lines of the UN or WHO

1 ( +1 / -0 )

Lots of these people losing their mind over AI,maybe people should embrace critical thinking,I make decisions everyday using the laws of probability,it give me statically base to achieve my goals

0 ( +1 / -1 )

People not thinking for themselves are more of a threat than AI,let others think for them as Zombies,

0 ( +1 / -1 )

Weak minded and the gullible people will always be suseptible to being manipulated by anything ,AI is just the other mean to do it

-1 ( +0 / -1 )

Every technology gets exploited for evil or money by governments, companies and people.

(Disclaimer: this comment was not produced by ChatGPT.)

0 ( +0 / -0 )

Login to leave a comment

Facebook users

Use your Facebook account to login or register with JapanToday. By doing so, you will also receive an email inviting you to receive our news alerts.

Facebook Connect

Login with your JapanToday account

User registration

Articles, Offers & Useful Resources

A mix of what's trending on our other sites